week 4 video research

Perception, 1999, volume 28, pages 1059 ̂ 1074

Gorillas in our midst: sustained inattentional blindness for dynamic events

Daniel J Simons, Christopher F Chabris Department of Psychology, Harvard University, 33 Kirkland Street, Cambridge, MA 02138, USA; e-mail: [email protected] Received 9 May 1999, in revised form 20 June 1999

Abstract. With each eye fixation, we experience a richly detailed visual world. Yet recent work on visual integration and change direction reveals that we are surprisingly unaware of the details of our environment from one view to the next: we often do not detect large changes to objects and scenes ( c̀hange blindness'). Furthermore, without attention, we may not even perceive objects (`inattentional blindness'). Taken together, these findings suggest that we perceive and remember only those objects and details that receive focused attention. In this paper, we briefly review and discuss evidence for these cognitive forms of `blindness'. We then present a new study that builds on classic studies of divided visual attention to examine inattentional blindness for complex objects and events in dynamic scenes. Our results suggest that the likelihood of noticing an unexpected object depends on the similarity of that object to other objects in the display and on how difficult the priming monitoring task is. Interestingly, spatial proximity of the critical unattended object to attended locations does not appear to affect detection, suggesting that observers attend to objects and events, not spatial positions. We discuss the implications of these results for visual representations and awareness of our visual environment.

1 Introduction

`̀ It is a well-known phenomenon that we do not notice anything happening in our surround- ings while being absorbed in the inspection of something; focusing our attention on a certain object may happen to such an extent that we cannot perceive other objects placed in the peripheral parts of our visual field, although the light rays they emit arrive completely at the visual sphere of the cerebral cortex.''

Rezso« Bä lint 1907 (translated in Husain and Stein 1988, page 91)

Perhaps you have had the following experience: you are searching for an open seat in a crowded movie theater. After scanning for several minutes, you eventually spot one and sit down. The next day, your friends ask why you ignored them at the theater. They were waving at you, and you looked right at them but did not see them. Just as we sometimes overlook our friends in a crowded room, we occasionally fail to notice changes to the appearance of those around us. We have all had the embarrassing experience of failing to notice when a friend or colleague shaves off a beard, gets a haircut, or starts wearing contact lenses. We feel that we perceive and remember everything around us, and we take the occasional blindness to visual details to be an unusual exception. The richness of our visual experience leads us to believe that our visual representations will include and preserve the same amount of detail (Levin et al 2000).

The disparity between the richness of our experience and the details of our repre- sentation, though `well known' to Bä lint in 1907, has been studied only sporadically in the psychological literature since then, and many of the most striking results appear to have been neglected by contemporary researchers. Although the past 20 years have seen increasing interest in the issue of the precision of visual representations, a series of studies from the 1970s and 1980s using dynamic visual displays provides some of the most dramatic demonstrations of the importance of attention in perception (see Neisser 1979 for an overview). In these studies, observers engage in a continuous task that requires them to focus on one aspect of a dynamic scene while ignoring others.

1060 D J Simons, C F Chabris

At some point during the task an unexpected event occurs, but the majority of observers do not report seeing it even though it is clearly visible to observers not engaged in the concurrent task (Becklen and Cervone 1983; Littman and Becklen 1976; Neisser 1979; Neisser and Becklen 1975; Rooney et al 1981; Stoffregen et al 1993; Stoffregen and Becklen 1989). Although these studies have profound implications for our understand- ing of perception with and without attention, and despite their obvious connection to more recent work on visual attention (eg change blindness, attentional blink, repetition blindness, inattentional blindness), the empirical approach has fallen into disuse. One goal of our research is to revive the approach used in these original studies of `selective looking' in the context of more recent work on visual attention.

Over the past few years, several researchers have demonstrated that conscious per- ception seems to require attention. When attention is diverted to another object or task, observers often fail to perceive an unexpected object, even if it appears at fixation ö a phenomenon termed `inattentional blindness' (eg Mack and Rock 1998).(1) These findings are reminiscent of another set of findings falling under the rubric of c̀hange blindness'. Observers often fail to notice large changes to objects or scenes from one view to the next, particularly if those objects are not the center of interest in the scene (Rensink et al 1997). For example, observers often do not notice when two people in a photograph exchange heads, provided that the change occurs during an eye movement (Grimes 1996; see Simons and Levin 1997 for a review). Such studies suggest that attention is necessary for change detection (see also Scholl 2000), but not sufficient, as even changes to attended objects are often not noticed (Levin and Simons 1997; Simons and Levin 1997, 1998; Williams and Simons 2000). For example, observers who were giving directions to an experimenter often did not notice that the experimenter was replaced by a different person during an interruption caused by a door being carried between them (Simons and Levin 1998).

Both areas of research focus on two fundamental questions. (i) To what degree are the details of our visual world perceived and represented? (ii) What role does attention play in this process? We will review recent evidence for inattentional blindness to provide a current context for a discussion of earlier research on the perception of unexpected events. We then present a new study examining the variables that affect inattentional blindness in naturalistic, dynamic events, and consider the results within the broader framework or recent attention research, including change blindness.

1.1 Inattentional blindness Studies of change blindness assume that, with attention, features can be encoded (abstractly or otherwise) and retained in memory. That is, all of the information in the visual environment is potentially available for attentive processing. Yet, without atten- tion, not much of this information is retained across views. Studies of inattentional blindness have made an even stronger claim: that, without attention, visual features of our environment are not perceived at all (or at least not consciously perceived)ö observers may fail not just at change detection, but at perception as well.

Recent work on the role of attention in perception has explored what happens to unattended parts of simple visual displays (Mack and Rock 1998; Mack et al 1992; Moore and Egeth 1997; Newby and Rock 1998; Rock et al 1992; Rubin and Hua 1998; Silverman and Mack 1997). In traditional models of visual search, features are often assumed to be processed preattentively if search speeds are unaffected by the number (1) Mack and Rock (1998) draw a distinction between conscious perception and implicit perception. Consistently with this distinction, when we use the term `perceive' (or `notice' or `see') in this paper, we mean that observers have at some point had a conscious experience of an object or event. How- ever, it is important to note that even when observers do not perceive an object, it may still have an implicit influence on their subsequent decisions and performance (eg Chun and Jiang 1998; Moore and Egeth 1997).

Gorillas in our midst 1061

of distracter items in the display (ie the feature `pops out' effortlessly). Preattentive processing of some features would allow for rapid perception of more complex objects that are built by combining such sensory primitives. However, visual search tasks may not truly assess the processing of unattended stimuli because observers have the expect- ation that a target may appearöobservers know that they will have to search the display for a particular stimulus. Hence, they may expect to perceive these features, which would allow their visual/cognitive system to anticipate the features. The inatten- tional-blindness paradigm developed by Mack, Rock, and colleagues avoids this potential confound of knowledge of the task (eg Mack and Rock 1998), allowing a more direct assessment of the perception of unattended stimuli. In a typical version of their task, observers judge which of two arms of a briefly displayed large cross is longer. On the fourth trial of this task, an unexpected object appears at the same time as the cross. After this trial, observers are asked to report if they saw anything other than the cross. After answering this question, observers view another trial, now with the suggestion that something might appear. This allows an assessment of perception under conditions of divided attention. Last, subjects complete a final, full-attention trial in which they look for and report the critical object but ignore the cross. Performance on the critical, unattended trial is compared with that on the divided-attention and full-attention trials to estimate the degree to which attention influences perception. The difference in the proportion of subjects noticing on the full-attention and critical trials is the amount of inattentional blindness.

Several clear patterns emerge from this body of research (see Mack and Rock 1998 for an overview). (i) About 25% of subjects are inattentionally blind when the cross is presented at fixation and the unexpected object is presented parafoveally (subjects typically detect the critical stimulus on divided-attention and full-attention trials). (ii) About 75% of subjects are inattentionally blind when the cross is presented para- foveally and the unexpected object is presented at fixation, suggesting an effortful shift of attention away from fixation to the cross and possible inhibition of processing at the ignored fixation location. (iii) These levels of detection are no different for features thought to be preattentively processed (eg color, orientation, motion) and those thought to require effort. (iv) Although objects composed of simple visual features are not easily detected, some meaningful stimuli are. Observers typically notice their own name or a smiley face even when they did not expect it. Note, however, they do not tend to notice their own name if one letter is changed (see also Rubin and Hua 1998). Observers do not consciously perceive the visual features, but they do perceive the meaning. (v) Observers seem to focus attention on particular locations on the screen. Objects that appear inside this zone of attention are more likely to be detected than those appearing out- side (Mack and Rock 1998; Newby and Rock 1998), suggesting that attention is focused not on the object or event itself, but on the area around that object.

1.2 `Selective looking' These recent studies of inattentional blindness used simple, brief visual displays under precisely controlled timing conditions, in the vein of work on visual search and related attention paradigms that were largely designed to examine how we select and process features and objects. The paradigm was designed to be a visual analogue of dichotic- listening studies conducted during the 1950s and 1960s (Cherry 1953; Moray 1959; Treisman 1964), and largely succeeded in replicating the classic auditory effects with visual stimuli. Although relatively little unattended information reaches awareness, some particularly meaningful stimuli do. Despite the similarity of these theoretical conclusions, they are fundamentally different in an important way. Almost by necessity, dichotic-listening tasks involve dynamic rather than static events. Listening studies reveal a degree of `inattentional deafness' that extends over time and over changes in the unattended stimulus.

1062 D J Simons, C F Chabris

In that sense, the computer-based inattention paradigm is not a true analogue of dichotic- listening tasks. Although the theoretical conclusions match our experience of not seeing friends in a crowded theater (and hearing our own name spoken at a noisy party), the experimental paradigm may not fully capture all aspects of that natural situation [see Neumann et al (1986) for a discussion of the difficulties of equating auditory and visual divided-attention tasks]. However, an earlier series of studies by Neisser and his colleagues did use dynamic events to address many of the same questions.

In an initial study (Neisser and Becklen 1975), observers viewed a display which presented two overlapping, simultaneous events. (The superimposition was achieved by showing both of the separately recorded events on an angled, half-silvered mirror.) One of the events was a hand-slapping game in which one player extended his hands with palms up and the other player placed his hands on his opponents hands with palms down. The player with palms up tries to slap the back of the other player's hands, and the other player tries to avoid the slap. The second event depicted three people moving in irregular patterns and passing a basketball. Subjects were asked to closely monitor one of the two events. If they monitored the hand game, they pressed a button with each attempted slap. If they monitored the ball game, they pressed the button for each pass. Each subject viewed a total of ten trials. The first two trials showed each of the games alone. On the 3rd and 4th trials, both events were presented simultaneously, but subjects were asked to follow only one of them. On the 5th and 6th trials, subjects attempted to respond to both events, using one hand to respond to each (only twenty actions per minute rather than forty occurred in these two and subsequent trials). On the last four trials, subjects responded to only one of the events, but an additional unexpected event occurred as well. In trial 7, the two hand-game players stopped and shook hands. On trial 8, one of the ball-game players threw the ball out of the game and the players continued to pretend to be passing the ball. The ball was returned after 20 s of fake throws. On trial 9, the hand-game players briefly stopped their game and passed a small ball back and forth. On trial 10, each of the ball-game players stepped off camera and was replaced by a woman and, after 20 s, the original men returned in the same fashion.

The results of this study are largely consistent with the findings of computer-based inattention studies. In the initial trials, subjects could easily follow one event while ignoring another event occupying the same spatial position. [This was true even when subjects were not allowed to move their eyes; see Littman and Becklen (1976).] Not surprisingly, they had much greater difficulty simultaneously monitoring both events. More importantly, in the initial trial with an unexpected event, only one of twenty-four people spontaneously reported the hand shake, and three others mentioned it in post- experiment questioning. None of the subjects spontaneously reported the disappearance of the ball, three spontaneously reported the ball pass in the hand game, and three reported the exchange of women for men on the final trial. Subjects who noticed one of the unusual events were more likely to notice subsequent unusual events, much as subjects in the divided-attention conditions in inattentional-blindness studies typically reported the presence of the previously `unexpected' object (Mack and Rock 1998). In total, 50% of Neisser and Becklen's (1975) subjects showed no indication of having seen any of the unexpected events, and even subjects who did notice typically could not accurately report the details of them.

In a more recent version of this sort of divided-visual-attention task, observers viewed superimposed videotapes of two of the ball games described above (Becklen, Neisser, and Littman, discussed in Neisser 1979).(2) The players in one game wore

(2) Many of the `selective-looking' studies conducted by Neisser and his colleagues were never published in complete empirical reports. In such cases, as here, we have cited unpublished or in-preparation manuscripts on the basis of their descriptions in other, published materials.

Gorillas in our midst 1063

black shirts and the players in the other game wore white shirts. This change made the attended and ignored events more similar, and therefore more difficult to discrim- inate. Nevertheless, observers could successfully follow one game while ignoring the other even when both teams wore the same clothing (in fact, the same three players appeared in each video stream).

In subsequent studies of selective looking, Neisser and his colleagues used this `basket- ball-game' task [see Neisser (1979) for a description of several different versions]. In the most famous demonstration, observers attend to one team of players, pressing a key whenever one of them makes a pass, while ignoring the actions of the other team. After about 30 s, a woman carrying an open umbrella walks across the screen (this video was also superimposed on the others so all three events were partially transparent; see figure 1). She is visible for approximately 4 s before walking off the far end of the screen. The games then continue for another 25 s before the tape is stopped. Of twenty-eight naive observers, only six reported the presence of the umbrella woman, even when questioned directly after the task (Neisser and Dube, cited in Neisser 1979). Interestingly, when subjects had practice performing the task on two similar trials before the trial with the unexpected umbrella woman, 48% noticed her. When subjects just watched the screen and did not perform any task, they always noticed the umbrella woman, a result consistent with the inattentional-blindness findings reviewed earlier (and with work on saccade-contingent changes; see Grimes 1996; McConkie and Zola 1979).

Figure 1. A single frame captured from a late-generation video of the umbrella-woman sequence used by Neisser and colleagues (eg Neisser 1979). The woman is in the center of the image and her umbrella is white.

Interestingly, Neisser (1979) mentioned an additional study in which the umbrella woman wore the same-color shirt as either the attended or the unattended team. Apparently, this feature-similarity manipulation caused little difference in the rate of noticing. Also, when the unexpected character was a small boy rather than the umbrella woman, fewer subjects noticed him, and when the umbrella woman stopped her motion and performed a little dance, more subjects noticed. These latter two findings suggest that properties of an unattended stimulus can capture attention, even though similarity to attended stimuli seemed to make little difference. However, these findings must be evaluated tentatively, because the details of the experimental paradigm were not presented by Neisser (1979).

1064 D J Simons, C F Chabris

In one of the few published empirical reports with this paradigm, Becklen and Cervone (1983) examined the effect of eliminating the delay between the umbrella- woman's appearance and the questioning of subjects. They found no difference in noticing rates when the entire video was shown (35%) and when the video ended immediately after the umbrella woman left the scene (33%). Furthermore, performance was substantially worse when the video ended as the umbrella woman was halfway across the court (7%), even though that meant that the last image subjects saw included the umbrella woman. These subjects provided accurate descriptions of the scene, including details of the player locations, but did not mention an umbrella woman. Neisser and Rooney (cited in Becklen and Cervone 1983) addressed the same question by interrupting the action as the umbrella woman was two-thirds of the way across the screen. Immediately after the interruption, a split screen appeared with the umbrella woman on one side and a boy holding a soda can on the other; subjects were asked to pick which they had seen. When the results were corrected for guessing, only 30% of observers had noticed the umbrella woman, a level comparable to that shown in postexperiment interviews.

These findings provide important evidence against the notion of `inattentional amnesia', an alternative account of findings of inattentional blindness and change blindness. According to this view (Wolfe 1999), the unexpected event is consciously perceived, but immediately forgotten. Hence, the failure to report its appearance reflects a failure of memory rather than of perception. In this case, however, even though subjects are tested immediately after the event, they are no better at detecting it. Further- more, when people notice the unexpected event in this task, they sometimes smile or laugh; nonnoticers show no outward signs of detection. The forgetting would have to be so rapid as to be inseparable from the act of perception to allow any sort of amnesia to explain these findings.

This early work on selective looking raised a number of questions needing further study. What role does similarity between the unexpected and attended events play in detection? Are particularly unusual events more or less likely to be detected? Does task difficulty increase or decrease detection? Perhaps the most important question left unanswered in this early work is what role the unusual superimposition of the events played in causing inattentional blindness. Most cognitive psychologists we have talked to found these results interesting, but were somewhat less convinced of the importance of the failures to notice unexpected events. After all, the video superimposition gives the displays an odd appearance, one not typically experienced in the real world and one in which the players and the umbrella woman are not as easy to see as they would be without superimposition.

One more recent study has looked at performance when all of the actors and the umbrella woman are shot from a single video camera, with no superimposition (Stoffregen et al 1993). Under these conditions, the players and umbrella woman occluded each other and the balls. If failures to notice the umbrella woman in earlier studies resulted from the unnatural appearance of the superimposed version of the display, performance might be much better with a `live' version. Subjects performed the task for approximately 30 s before the umbrella woman appeared and walked across the screen. The camera angle used for this film was much wider than in earlier studiesöit showed an entire regulation basketball court. Consequently, the umbrella woman was visible for a longer time (12 s) and the players and the umbrella woman were substantially smaller on screen than in earlier studies. Another notable difference is that only twelve passes occurred during the 60-s video (rather than 20 ̂ 40 as in earlier studies). Even in this live version of the study, only three of twenty normal subjects tested reported the presence of the umbrella woman. Although this finding does suggest that visual superimposition was not the cause of failures of noticing, it did

Gorillas in our midst 1065

not match the stimulus conditions of the other studies and did not directly compare performance with and without superimposition. The difference in camera angle (and consequent character size) alone may well have affected detection rates, so this study is not a well-controlled test of the generalizability of inattentional-blindness phenomena to more natural stimulus conditions.

Despite the importance of all the unanswered questions raised by these studies, to our knowledge the findings reviewed above are the only published reports using dynamic, naturalistic events to study the detection of unexpected objects.(3) Taken together, these studies lead to a number of striking conclusions, some consistent and others inconsis- tent with findings with simple displays. Unlike the computer-based studies (eg Mack and Rock 1998; Newby and Rock 1998), the video studies demonstrate that inatten- tional blindness does not result from attention being focused elsewhere in the display. In the superimposed version of the display, the umbrella woman occupied exactly the same spatial position as the attended players and balls. In fact, the balls even passed through the umbrella woman. This finding is inconsistent with the computer-based result that detection was better when the unexpected object appeared within the region defined by the attended object (Mack and Rock 1998). Several factors might account for this difference. First, there were simply more objects to attend to in the video displays, so attention may not have stayed on any one location for long. Second, the dynamic display may have captured and held attention more effectively than the cross task. Third, the video task may simply have been harder, leaving fewer attentional resources available to process unanticipated events. These video studies do show that a form of inattentional blindness can last much longer than the brief exposure times used in recent static-display studies. Subjects missed ongoing events that lasted for more than 4 s.

Although these differences between the computer-based and video studies are important, the general similarity of the conclusions is striking. In both cases, observers often do not see unanticipated objects and events. The video studies suggest that these findings can help explain real-world phenomena such as our inability to see our friends in a crowded movie theater or airplanes on an approaching runway when our attention is focused on a different goal. Both change blindness and inattentional blindness show that attention plays a critical role in perception and in representation. Without atten- tion, we often do not see unanticipated events, and even with attention, we cannot encode and retain all the details of what we see.

Although these video studies of inattentional blindness help to generalize findings from simple displays to more complex situations, the original reports do not fully examine all of the critical questions. For example, there is a hint that the visual similarity of the unexpected object to the attended ones makes no difference, but the details of that study were never published. Furthermore, the experiments did not systematically consider the role of task difficulty in detection. Perhaps most importantly, no direct comparisons were made between performance with the superimposed version of the display and with the `live' version. In the studies reported here, we attempt to examine each of these factors. We also consider the nature of the unusual event. To combine all of these factors orthogonally within a single consistent paradigm, we filmed several video segments with the same set of actors in the same location on the same day. We then asked a large number of naive observers to watch the video recordings and later answer questions about the unexpected events.

(3) Haines (1989) did address this topic as part of a larger human-interface study. Pilots attempted to land a plane in a flight simulator while using a head-up display of critical flight information superimposed on the `windshield'. Under these conditions, some pilots failed to notice that a plane on the ground was blocking their path. In addition, Mack and Rock (1998) report several studies in which the unexpected object moved stroboscopically across part of the display, often without being detected during the 200 ms viewing period.

1066 D J Simons, C F Chabris

2 Method 2.1 Observers 228 observers, almost all undergraduate students, participated in the experiment. Each observer either volunteered to participate without compensation, received a large candy bar for participating, or was paid a single fee for participating in a larger testing session including another, unrelated experiment.

2.2 Materials Four videotapes, each 75 s in duration, were created. Each tape showed two teams of three players, one team wearing white shirts and the other wearing black shirts, who moved around in a relatively random fashion in an open area (approximately 3 m deep65.2 m wide) in front of a bank of three elevator doors. The members of each team passed a standard orange basketball to one another in a regular order: player 1 would pass to player 2, who would pass to player 3, who would pass to player 1, and so on. The passes were either bounce passes or aerial passes; players would also dribble the ball, wave their arms, and make other movements consistent with their overall pattern of action, only incidentally looking directly at the camera.

After 44 ̂ 48 s of this action, either of two unexpected events occurred: in the Umbrella-Woman condition, a tall woman holding an open umbrella walked from off camera on one side of the action to the other, left to right. The actions of the players, and this unexpected event, were designed to mimic the stimuli used by Neisser and colleagues. In the Gorilla condition, a shorter woman wearing a gorilla costume that fully covered her body walked through the action in the same way. In either case, the unexpected event lasted 5 s, and the players continued their actions during and after the event.

There were two styles of video: in the Transparent condition, the white team, black team, and unexpected event were all filmed separately, and the three video streams were rendered partially transparent and then superimposed by using digital video-editing software. (Neisser and colleagues achieved similar effects using analog equipment or a physical apparatus that superimposed separate displays by means of mirrors.) In the Opaque condition, all seven actors were filmed simultaneously and could thus occlude one another and the basketballs; this required some rehearsal to eliminate collisions and other accidents, and to achieve natural-looking patterns of movement. All videos were filmed with an SVHS video camera (Panasonic AG456U) and were digitized and edited by using a nonlinear digital-editing system (Media 100LX and Adobe Aftereffects, running on Power Computing hardware). All editing of the videos was accomplished after digitization, so the degree of signal loss due to multiple generations of editing was minimized and also equated across conditions. Stimuli were created by mastering the digitally edited sequences to VHS format tapes. Thus, as shown in figure 2, videos were created with the following four display types: Transparent/Umbrella Woman, Transparent/Gorilla, Opaque/Umbrella Woman, and Opaque/Gorilla. The first of these was most similar to the conditions tested by Neisser and colleagues.

2.3 Procedures All observers were tested individually and gave informed consent in advance. Before viewing the videotape, observers were told that they would be watching two teams of three players passing basketballs and that they should pay attention to either the team in white (the White condition) or the team in black (the Black condition). They were told that they should keep either a silent mental count of the total number of passes made by the attended team (the Easy condition) or separate silent mental counts of the number of bounce passes and aerial passes made by the attended team (the Hard condition). Thus, for each of the four displays, there were four task conditions öWhite/Easy,White/Hard, Black/Easy, and Black/Hardöfor a total of sixteen individual conditions. Each observer participated in only one condition.

Gorillas in our midst 1067

Transparent/Umbrella Woman Transparent/Gorilla

Opaque/Umbrella Woman Opaque/Gorilla

Figure 2. Single frames from each of the display tapes used here. (These tapes and that referred to in figure 3 were in color. These frames are displayed in color on http://www.perceptionweb/ perc0999/simons.htm/ and archived on the CD ROM supplied with issue 12 of Perception.) The transparent conditions (top row) were created by superimposing three separately filmed events by means of digital video editing. The opaque conditions (bottom row) were filmed as a single action sequence with all seven actors. This figure shows the display for each condition halfway through the unexpected event, which lasted for 5 s of the 75-s-long video.

After viewing the videotape and performing the monitoring task, observers were immediately asked to write down their count(s) on paper.(4) They were then asked to provide answers to a surprise series of additional questions. (i) While you were doing the counting, did you notice anything unusual on the video? (ii) Did you notice any- thing other than the six players? (iii) Did you see anyone else (besides the six players) appear on the video? (iv) Did you see a gorilla [woman carrying an umbrella] walk across the screen? After any “yes'' response, observers were asked to provide details of what they noticed. If at any point an observer mentioned the unexpected event, the remaining questions were skipped. After the questioning, observers were asked whether they had ever previously participated in an experiment similar to this or had ever heard of such an experiment or the general phenomenon. (Observers who answered `̀ yes'' were replaced and their data were discarded.) Last, the observer was debriefed; this included replaying the videotape on request. Each testing session lasted 5 ̂ 10 min.

Twenty-one experimenters tested the observers. To ensure uniformity of procedures, we developed a written protocol in advance and reviewed it with the experimenters before they began to collect data. This document specified what the experimenters would say to (4) Note that in all the Transparent conditions, the correct counts were identical because the same passing sequences were used to create both of the Transparent display tapes (Umbrella Woman and Gorilla). In the Opaque conditions, the correct counts varied because the passing sequences were filmed separately for each of the unexpected events.

1068 D J Simons, C F Chabris

each observer, when they would say it, how and when they would show the videotape, how they would collect and record the data, and how they would debrief observers. Experimenters used a variety of television monitors, ranging from 13 to 36 inches (diagonal) in screen size to present the videotapes.

3 Results Data from thirty-six observers were discarded for a variety of reasons: either (i) the observer already knew about the phenomenon and/or experimental paradigm (n ‹ 14), (ii) the observer reported losing count of the passes ( n ‹ 9), (iii) passes were incompletely or inaccurately recorded (n ‹ 7), (iv) the observer's answer could not be clearly inter- preted (n ‹ 5), or (v) the observer's total pass count was more than three standard deviations away from the mean of the other observers in that condition (n ‹ 1). The remaining 192 observers were distributed equally across the sixteen conditions of the 2626262 design (twelve per condition).

Although we asked a series of questions escalating in specificity to determine whether observers had noticed the unexpected event, only one observer who failed to report the event in response to the first question (`̀ did you notice anything unusual?'') reported the event in response to any of the next three questions (which culminated in `̀ did you see a … walk across the screen?''). Thus, since the responses were nearly always consistent across all four questions, we will present the results in terms of overall rates of noticing. Table 1 shows these results for each of the sixteen conditions.

Table 1. Percentage of subjects noticing the unexpected event in each condition. Each row corre- sponds to one of the four video display types. Columns are grouped by monitoring task and attended team (White or Black). In the Easy task, subjects counted the total number of passes made by the attended team. In the Hard task, subjects maintained separate simultaneous counts of the aerial and bounce passes made by the attended team.

Easy task Hard task

White team Black team White team Black team

Transparent Umbrella Woman Gorilla

Opaque Umbrella Woman Gorilla

58 8

100 42

92 67

58 83

33 8

83 50

42 25

58 58

Out of all 192 observers across all conditions, 54% noticed the unexpected event and 46% failed to notice the unexpected event, revealing a substantial level of sustained inattentional blindness for a dynamic event and confirming the basic results of Neisser and colleagues. More observers noticed the unexpected event in the Opaque condition (67%) than in the Transparent condition (42%); w1

2 ‹ 12:084, p 5 0:001; n ‹ 96 per condition. However, even in the Opaque case, a substantial proportion of observers (33%) failed to report the event, despite its visibility and the repeated questions about it.

More observers noticed the unexpected event in the Easy (64%) than in the Hard (45%) condition (w1

2 ‹ 6:797, p 5 0:009; n ‹ 96 per condition). To confirm that these monitoring tasks differed in difficulty, we calculated the SD of the total pass counts reported by observers in each condition; the average SD was 2.71 in the eight Easy conditions and 6.77 in the eight Hard conditions, indicating that the Hard moni- toring task was indeed more difficult. Accordingly, the correlation across conditions between the frequency of noticing (shown in table 1) and the SD of the total pass count was r ‹ÿ 0:56. The effect of task difficulty was greater in the Transparent con- ditions (Easy 56%, Hard 27%; w1

2 ‹ 8:400, p 5 0:004; n ‹ 48 per condition) than in

Gorillas in our midst 1069

the Opaque conditions (Easy 71%, Hard 62%; w 21 ‹ 0:750, p 5 0:386; n ‹ 48 per con- dition), suggesting a multiplicative effect on residual attention capacity of tracking difficult-to-see stimuli and keeping two running counts in working memory.

Next we examined differences in the detection of the two unexpected events. The Umbrella Woman was noticed more often than the Gorilla overall (65% versus 44%; w 21 ‹ 8:392, p 5 0:004; n ‹ 96 per condition). This relation held regardless of the video type, monitoring task, or attended team, suggesting that the Umbrella Woman was either a more visually salient event than the Gorilla,(5) more consistent with observers' expectations about situations involving basketballs, more semantically similar to the attended events, or all three. However, when observers attended to the actions of the Black team, they noticed the Gorilla much more often than when they attended to the actions of the White team (Black 58%, White 27%; w1

2 ‹ 9:579, p 5 0:002; n ‹ 48 per condition). By contrast, attending the Black team versus the White team made little difference in noticing the Umbrella Woman (Black 62%, White 69%; w 21 ‹ 0:416, p 5 0:519; n ‹ 48 per condition). The Gorilla was black, whereas the Umbrella Woman wore pale colors that differed from both the Black and the White team. Thus, contrary to the suggestion of Neisser (1979), it appears that observers are more likely to notice an unexpected event that shares basic visual featuresöin this case, coloröwith the events they are attending to. In a sense, this effect is the opposite of the traditional `pop-out' phenomenon in visual search tasks, which occurs when an item that differs in basic visual features from the rest of the display is easier to notice and identify.

It is possible that subjects who lost count of the passes would be most likely to notice the unexpected event. This is unlikely, however, for two reasons. First, subjects who reported losing count were replaced prior to data analysis. Second, we calculated the point-biserial correlation r between noticing (coded as 1 for reporting and 0 for not reporting the event) and the subject's absolute deviation from an accurate pass count (measured as the number of passes above or below the correct range) for each condi- tion except the Opaque/Umbrella-Woman/White/Easy condition, which engendered 100% noticing. Across these fifteen conditions the correlations averaged to r ‹ 0:15, suggesting that noticing was not strongly associated with counting poorly or inattentively.

4 Discussion Our findings have replicated, generalized, and extended the surprising result first reported by Neisser and colleagues (Bahrick et al 1981; Becklen and Cervone 1983; Littman and Becklen 1976; Neisser 1979; Neisser and Becklen 1975; Rooney et al 1981; Stoffregen et al 1993; Stoffregen and Becklen 1989), and have demonstrated a robust phenomenon of sustained inattentional blindness for dynamic events. In particular, we have shown the following. (i) Approximately half of observers fail to notice an ongoing and highly salient but unexpected event while they are engaged in a primary monitoring task. This extends the phenomenon of inattentional blindness (eg Mack and Rock 1998) by at least an order of magnitude in the duration of the event that can be missed. To stretch this limit further, we tested a longer and more salient unexpected event in an additional condi- tion not reported above. In a separate Opaque-style video recording, the Gorilla walked from right to left into the live basketball-passing event, stopped in the middle of the players as the action continued all around it, turned to face the camera, thumped its chest, and then resumed walking across the screen (this action began after 35 s and lasted 9 s in a stimulus tape 62 s long; see figure 3 for a still frame).

(5) Visual salience, here, could refer to the relative distinctiveness of the unexpected objects in relation to the other players or to the background of the scene. Furthermore, the Umbrella Woman may have been spatially more distinctive in that her umbrella extended above the heads of the other players whereas the Gorilla was the same height as the other players.

1070 D J Simons, C F Chabris

Figure 3. A single frame from an additional experimental condition in which the gorilla stopped in the middle of the display, turned to face the camera, thumped its chest, and then continued walking across the field of view. Subjects performed the Easy monitoring task while attending to the White team, and the noticing rate was similar to that in the corresponding condition with the standard (shorter) Opaque/Gorilla event.

Twelve new observers (6) watched this video while attending to the White team and engaging in the Easy monitoring task; only 50% noticed the event. This is roughly the same as the percentage that noticed the normal Opaque/Gorilla-walking event (42%) under the same task conditions. (ii) This sustained inattentional blindness occurs more frequently if the display is trans- parent, with actors seeming to move through each other (as used in earlier studies), but observers often miss even fully visible objects appearing in live-action opaque dis- plays. This latter finding is contrary to the intuitions of researchers who believed that the original effect was due to the unusual nature of the transparent video, and provides further evidence that inattentional blindness is a ubiquitous perceptual phenomenon rather than an artifact of particular display conditions. (iii) The level of inattentional blindness depends on the difficulty of the primary task; in principle, inattentional blindness in this paradigm could be continuously varied by appropriately manipulating the difficulty of the monitoring task. (iv) Observers are more likely to notice unexpected events if these events are visually similar to the events they are paying attention to. (On the basis of our results it is logically possible that dissimilarity to the ignored events is instead the crucial factor.) (v) Objects can pass through the spatial extent of attentional focus (and the fovea) and still not be `seen' if they are not specifically being attended. This conclusion is consistent with Mack and Rock's (1998) finding that observers often fail to notice a bar or square moving stroboscopically across fixation during a 200 ms display. In each of our videotapes, the unexpected object more than once crossed the path of the basketball and/or a player throwing or catching the ball, the observers would have had to pay attention to both of those elements of the display to perform the monitoring task.

In most respects, the results of this study are consistent with computer-based studies of inattentional blindness. Observers fail to report unexpected, suprathreshold objects when they are engaged in another task. Both sets of findings are consistent with the (6) Data from two additional observers were discarded, one because he already knew about the effect, the other because his answer could not be clearly interpreted.

Gorillas in our midst 1071

claim that there is no conscious perception without attention. The consistency of the theoretical conclusions that can be drawn from these two radically different paradigms is reassuring. Whether the unexpected object is flashed for 200 ms in an otherwise empty display or it moves dynamically across a natural scene for 5 s, observers are unlikely to notice it if attention is otherwise engaged. This consistency suggests that the results of computer-based studies of inattentional blindness can and do generalize to situations closer to our real-world experiences.

The results of our experiments also call to mind recent findings from research on change blindness. Many studies of change blindness focus on simple displays of letters or dots to determine how little information is preserved from one view to the next. More recently, change-blindness research has moved from using simple displays of letters, dots, and words (eg Pashler 1988; Phillips 1974) to more complex, naturalistic displays for which more information is available for selection (see Simons and Levin 1997 for a review). Given the simplicity and relative meaninglessness of the simple displays, the generalizability of the results to more naturalistic viewing conditions was not certain (see Simons 2000, for discussion). The recent thrust of work on change blindness has been to examine whether the inferences drawn from work with simple displays will hold for more natural displays. One dramatic demonstration was at least partly responsible for this move toward increased naturalism. When viewing photographs of natural scenes in preparation for a memory test, people missed large, meaningful changes that occurred during eye movements (Grimes 1996). For example, observers often failed to notice when two people in a photograph exchanged hats or even when they exchanged heads. These findings have been replicated in subsequent work on saccade-contingent changes (Currie et al 1995; Henderson and Hollingworth, in press; McConkie and Currie 1996).

This change blindness for natural scenes has been extended to a number of other paradigms. For example, when an original and modified image are presented in rapid alternation with a blank screen interposed between them, observers have great difficulty detecting changes (Rensink et al 1997). This `flicker' technique shows that change blind- ness is not limited to saccade-contingent changes. In the case of saccade-contingent changes, the blur on the retina caused by the eye movement itself leads to suppres- sion of visual processing during the change, thereby preventing detection of any local transients. The blank screen in the flicker study has essentially the same effect, producing a global change signal that prevents detection of the local one caused by the change. Similar change blindness has been shown for changes across cuts or pans in motion pictures (Levin and Simons 1997; Simons 1996), eye blinks (O'Regan et al 2000), and `mud splashes' (O'Regan et al 1999). As noted earlier, even when one conversation partner is replaced by a different person during a brief interruption, observers often fail to notice the change (Simons and Levin 1998).

As in studies of inattentional blindness, the likelihood of change detection depends on the focus of attention. In studies of inattentional blindness, when observers are attending to another object or event, they are less likely to notice the unexpected event. In studies of change detection, people are better able to report changes to attended than unattended objects. For example, people are faster to detect changes in the flicker paradigm when the changed object is of central interest in the scene (Rensink et al 1997). Central objects are more likely to garner attentional resources, and if we have a limited capacity for holding information across views, changes to objects that receive more effortful processing are more likely to be detected (see Rensink 2000; Scholl 2000). Just as we often fail to perceive unexpected events, we often fail to notice unexpected changes to the visual details of our environmentöin both cases, this applies even when attention is focused on the area of the event or change.

Although the theoretical conclusions drawn from real-world studies are not altogether different from those derived from work with simple displays, they do show that change

1072 D J Simons, C F Chabris

blindness is a general property of the visual system and that it applies to almost all aspects of visual processing. We apparently do not retain a detailed visual representation of our surroundings from one view to the next, even for displays with all the richness of natural scenes. Similarly, studies of sustained inattentional blindness suggest that we fail to perceive unexpected objects even under naturalistic viewing conditions.

The results of our studies of sustained inattentional blindness, however, do contrast in an interesting way with those of one recent change-blindness study (Simons et al, in preparation). In that experiment, a female experimenter dressed in athletic clothing and carrying a basketball approaches a passerby in public and asks directions to a gym. During this interaction, a crowd of confederates walked between the two and surreptitiously took the basketball away. When asked if they noticed anything changed or anything different about her appearance, a minority of observers reported noticing that the basketball was gone. But when asked a follow-up question specifically referring to the basketball, most of the remaining observers `remembered' the basketball and were able to describe its unusual coloring. Thus, a visual change can be encoded but not explicitly reported until a specific retrieval cue is provided. In the experiment reported here, however, not one of the eighty-eight nonnoticers `remembered' the Gorilla or Umbrella-Woman events when specifically asked about it, and several did not believe that the event had happened until the videotape was replayed for them.

While there are several important differences between these paradigms that could account for this difference in behavior, they share the feature that a condition of inattention was created (by the conversation in the basketball disappearance study or by the monitoring task here) that apparently prevented many observers from becoming aware of a salient visual change. Perhaps the crucial difference is that whereas the conversation simply reduced the observer's attention by drawing it away from the critical object, the monitoring task in this study required observers to attend to one event while ignoring another that was happening in the same region of space. This `directed ignoring' could inhibit perception of not just the ignored event but of all unattended events, thereby preventing the formation of an explicit memory trace. Whether inattentional blindness occurs because the target is similar to the intentionally ignored items or different from the attended items is an open question that would be relatively difficult to explore by using video-based displays but could be explored by using more controlled computer-based tasks.

One alternative interpretation of our findings is that subjects did consciously perceive the unexpected object, however briefly, but immediately forgot they had seen it (Wolfe 1999). Although this inattentional-amnesia explanation can in principle account for our findings, it seems less plausible that the inattentional-blindness account for a number of reasons. First, detecting unusual objects or events would be a useful function for a visual system to have; immediately forgetting them would defeat this purpose. This is especially true for a prolonged, dynamic event. Given that the unexpected object in our experiments was available for further examination (something that was less true of earlier studies with briefly flashed objects), we might expect observers to try to verify their percept in these studies, thereby leading to a preserved representation. Furthermore, if observers did consciously perceive and then forget the gorilla, they presumably would not be particularly surprised when asked if there had been a gorilla in the display. Yet, observers in our study were consistently surprised when they viewed the display a second time, some even exclaiming, ̀ `I missed that?!'' It seems more parsimonious to assume that observers were never aware of the unexpected object than to assume that they saw a gorilla, then forgot about it, and then were shocked to see it when told to look for it. Last, as noted earlier, Becklen and Cervone (1983) found no improvement in noticing when the video was stopped immediately after the unexpected event rather than several seconds later. However, finding a direct test to distinguish between never perceiving an object and

Gorillas in our midst 1073

immediately forgetting it will be difficult because the inattentional-amnesia proponents could always argue that the memory test came too late. Thus, it may not be possible to distinguish empirically between the amnesia and the blindness explanations.

Although our findings suggest that unexpected events are often overlooked, the question of whether they leave an implicit trace remains open. Unnoticed stimuli in the static-inattentional-blindness paradigm can lead to priming effects (Mack and Rock 1998). However, those experiments did not require subjects to ignore anything. Neisser and colleagues found that subjects under the conditions we have described as directed ignoring were no more likely to select the unexpected object in a two-alternative forced-choice recognition test than were other subjects when asked to report it directly (Neisser and Rooney 1982, as cited in Becklen and Cervone 1983). However, forced choice may not be as sensitive as other implicit memory tests. Future research should explore the issues of preserved representations and directed ignoring within the sustained- inattentional-blindness paradigm we have reintroduced here.

Acknowledgements. Many thanks to all of the people who helped with filming the videos or collecting the data for this study: Jennifer Shephard (Umbrella Woman), Elisa Cheng (Gorilla), Judith Danovitch, Steve Most, Alex Wong (White team), Amy DeIpolyi, Jason Jay, Megan White (Black team), Dan Ellard, Samantha Glass, Jeremy Gray, Sara Greene, Annya Hernandez, Orville Jackson, Latanya James, David Marx, Steve Mitroff, Carolyn Racine, Kathy Richards, Chris Russell, Laurie Santos, Steve Stose, Ojas Tejani, Dan Tristan, Amy Wiseman, Leah Wittenberg, and Amir Zarrinpar (data collection, in alphabetical order). Thanks also to M J Wraga for suggesting the first part of the title of this article, to Brian Scholl for discovering the Bä lint quotation, to Dick Neisser for inspiration and for giving permission to produce figure 1, to Larry Taylor for helping to avoid collisions during the filming of the videos, and especially to Jerry Kagan for lending us his gorilla suit. Additional thanks to Steve Most, Brian Scholl, two anonymous reviewers, and everyone else who commented on earlier versions of this manuscript and presentations of these results. Miniaturized and abbreviated versions of the videos used in this study are available, in QuickTime

format, via the Internet at: http://coglab.wjh.harvard.edu/gorilla/index.html. No animals were harmed during the making of the videos.

References Bahrick L E, Walker A S, Neisser U, 1981 `̀ Selective looking by infants'' Cognitive Psychology 13

377 ̂ 390 Becklen R, Cervone D, 1983 `̀ Selective looking and the noticing of unexpected events'' Memory

and Cognition 11 601 ^ 608 Cherry E C, 1953 `̀ Some experiments upon the recognition of speech, with one and with two

ears'' Journal of the Acoustical Society of America 25 975 ̂ 979 Chun M M, Jiang Y, 1998 `̀ Contextual cueing: Implicit learning and memory of visual context

guides spatial attention'' Cognitive Psychology 36 28 ̂ 71 Currie C, McConkie G W, Carlson-Radvansky L A, Irwin D E, 1995 `̀ Maintaining visual stability

across saccades: Role of the saccade target object'' technical report UIUC-BI-HPP-95-01, Beckman Institute, University of Illinois

Grimes J, 1996 `̀ On the failure to detect changes in scenes across saccades'', in Perception (Vancouver Studies in Cognitive Science) Ed. K Akins, volume 2 (New York: Oxford University Press) pp 89 ^ 110

Haines R F, 1989 `̀A breakdown in simultaneous information processing'', in Presbyopia Research: From Molecular Biology to Visual Adaptation Eds G Obrecht, L W Stark (New York: Plenum) pp 171 ^ 175

Henderson J M, Hollingworth A, in press `̀ The role of fixation position in detecting scene changes across saccades'' Psychological Science

Husain M, Stein J, 1988 `̀ Rezso« Bä lint and his most celebrated case''Archives of Neurology 45 89 ̂ 93 Levin D T, Momen N, Drivdahl S B, Simons D J, 2000 `̀ Change blindness blindness: The meta-

cognitive error of overestimating change-detection ability'' Visual Cognition (in press) Levin D T, Simons D J, 1997 `̀ Failure to detect changes to attended objects in motion pictures''

Psychonomic Bulletin and Review 4 501 ̂ 506 Littman D, Becklen R, 1976 `̀ Selective looking with minimal eye movements'' Perception & Psycho-

physics 20 77 ^ 79 Mack A, Rock I, 1998 Inattentional Blindness (Cambridge, MA: MIT Press)

1074 D J Simons, C F Chabris

Mack A, Tang B, Tuma R, Kahn S, 1992 `̀ Perceptual organization and attention'' Cognitive Psychology 24 475 ̂501

McConkie G W, Currie C B, 1996 `̀ Visual stability across saccades while viewing complex pictures'' Journal of Experimental Psychology: Human Perception and Performance 22 563 ̂ 581

McConkie G W, Zola D, 1979 `̀ Is visual information integrated across successive fixations in reading?'' Perception & Psychophysics 25 221 ̂ 224

Moore C M, Egeth H, 1997 `̀ Perception without attention: Evidence of grouping under conditions of inattention'' Journal of Experimental Psychology: Human Perception and Performance 23 339 ̂352

Moray N, 1959 `̀Attention in dichotic listening: Affective cues and the influence of instructions'' Quarterly Journal of Experimental Psychology 11 56 ̂ 60

Neisser U, 1979 `̀ The control of information pickup in selective looking'', in Perception and its Development: A Tribute to Eleanor J Gibson Ed. A D Pick (Hillsdale, NJ: Lawrence Erlbaum Associates) pp 201 ̂ 219

Neisser U, Becklen R, 1975 `̀ Selective looking: Attending to visually specified events'' Cognitive Psychology 7 480 ̂ 494

Neumann O, Heijden A H C van der, Allport D A, 1986 `̀ Visual selective attention: Introductory remarks'' Psychological Research 48 185 ̂188

Newby E A, Rock I, 1998 `̀ Inattentional blindness as a function of proximity to the focus of attention'' Perception 27 1025 ̂ 1040

O'Regan J K, Deubel H, Clark J J, Rensink R A, 2000 `̀ Picture changes during blinks: Looking without seeing and seeing without looking'' Visual Cognition (in press)

O'Regan J K, Rensink R A, Clark J J, 1999 `̀ Change-blindness as a result of `mudsplashes' '' Nature (London) 398 34

Pashler H, 1988 `̀ Familiarity and visual change detection'' Perception & Psychophysics 44 369 ̂ 378 Phillips W A, 1974 `̀ On the distinction between sensory storage and short-term visual memory''

Perception & Psychophysics 16 283 ^ 290 Rensink R A, 2000 ̀ `Visual search for change: A probe into the nature of attentional processing'' Visual

Cognition (in press) Rensink R A, O'Regan J K, Clark J J, 1997 “To see or not to see: The need for attention to

perceive changes in scenes'' Psychological Science 8 368 ^ 373 Rock I, Linnett C M, Grant P, Mack A, 1992 `̀ Perception without attention: Results of a new

method'' Cognitive Psychology 24 502 ̂ 534 Rooney P, Boyce C, Neisser U, 1981 `̀A developmental study of noticing unexpected events''

unpublished manuscript Rubin N, Hua X L, 1998 `̀ Perceiving occluded objects under conditions of inattention'' Investigative

Ophthalmology & Visual Science 39(4) S1113 Scholl B J, 2000 `̀Attenuated change blindness for exogenously attended items in a flicker para-

digm'' Visual Cognition (in press) Silverman M, Mack A, 1997 `̀ Priming by iconic images'' Investigative Ophthalmology & Visual

Science 38(4) S963 Simons D J, 1996 `̀ In sight, out of mind: When object representations fail'' Psychological Science

7 301 ̂ 305 Simons D J, 2000 `̀ Current approaches to change blindness'' Visual Cognition (in press) Simons D J, Chabris C F, Levin D T, (in preparation) `̀ Change blindness is not caused by later

events overwriting earlier ones in visual short term memory'' Simons D J, Levin D T, 1997 `̀ Change blindness'' Trends in Cognitive Sciences 1 261 ̂ 267 Simons D J, Levin D T, 1998 `̀ Failure to detect changes to people in a real-world interaction''

Psychonomic Bulletin and Review 5 644 ̂ 649 Stoffregen T A, Baldwin C A, Flynn S B, 1993 `̀ Noticing of unexpected events by adults with

and without mental retardation'' American Journal on Mental Retardation 98 273 ̂ 284 Stoffregen T A, Becklen R C, 1989 “Dual attention to dynamically structured naturalistic events''

Perceptual and Motor Skills 69 1187 ̂1201 Treisman A, 1964 “Monitoring and storage of irrelevant messages in selective attention'' Journal

of Verbal Learning and Verbal Behavior 3 449 ̂ 459 Williams P, Simons D J, 2000 “Detecting changes in novel, complex three-dimensional objects''

Visual Cognition (in press) Wolfe J M, 1999 `̀ Inattentional amnesia'', in Fleeting Memories: Cognition of Brief Visual Stimuli

Ed. V Coltheart (Cambridge, MA: MIT Press) pp 71 ̂ 94

ß 1999 a Pion publication printed in Great Britain

RACIAL IDENTIFICATION AND PREFERENCE 169

5.

RACIAL IDENTIFICATION AND PREFERENCE I N NEGRO CHILDREN By Kenneth B. Clark and Mamie P. Clark

PROBLEM

The specific problem of this study is an analysis of the genesis and development of racial identification as a function of ego development and self-awareness in Negro children.

Race awareness, in a primary sense, is defined as a consciousness of the self as belonging to a specific group which is differentiated from other observable groups by obvious physical character­ istics which are generally accepted as being racial characteristics.

Because the problem of racial identi­ fication is so definitely related to the problem of the genesis of racial attitudes in children, it was thought practicable to attempt to determine the racial atti­ tudes or preferences of these Negro chil­ dren-and to define more precisely, as far as possible, the developmental pat­ tern of this relationship.

PROCEDURE

This paper presents results from only one of several techniques devised and used by the authors to investigate the development of racial identification and preferences in Negro children.1 Results presented here are from the Dolls Test.

Dolls Test. The subjects were pre­ sented with four dolls, identical in every respect save skin color. Two of these dolls were brown with black hair and two were white with yellow hair. In the ex­ perimental situation these dolls were un-

clothed except for white diapers. The position of the head, hands, and legs on all the dolls was the same. For half of the subjects the dolls were presented in the order: white, colored. white, colored. For the other half the order of presenta­ tion was reversed. In the experimental situation the subjects were asked to re­ spond to the following requests by choosing one of the dolls and giving it to the experimenter:

1. Give me the doll that you like to play with- (a) like best.

2. Give me the doll that is a nice doll. 3. Give me the doll that looks bad. 4. Give me the doll that is a nice color. 5. Give me the doll that looks like a

white child. 6. Give me the doll that looks like a

colored child. 7. Give me the doll that looks like a

Negro child. 8. Give me the doll that looks like

you. Requests 1 through 4 were designed

to reveal preferences; requests 5 through 7 to indicate a knowledge of "racial differences"; and request 8 to show self­ identification.

It was found necessary to present the preference requests first in the experi­ mental situation because in a preliminary investigation it was clear that the chil­ dren who had already identified them­ selves with the colored doll had a marked tendency to indicate a preference for this doll and this was not necessarily a gen-

Condensed by the authors from an unpublished study made possible by a fellowship grant from the Julius Rosenwald Fund, 1940-1941.

1 Other techniques presented in the larger study include: (1) a coloring test; (2) a questionnaire and (3) a modification of the Horowitz line drawing technique. (R. E. Horowitz, "Racial Aspects of Self-identification in Nursery School Children," J. Psychol., 1939, VII, 91-99.)

Age, years North South Total

3 13 18 31 4 10 19 29 5 34 12 46 6 33 39 72 7 29 46 75

Total . 119 134 253

Sex distribution of subjects:

___Sex ____ , ___ North South Total — – – –

Male . 53 63 116 Female 66 71 137

Skin color of subjects:

Skin color North South Total

Light,. 33 13 46 Medium b 58 70 128 Dark 0 28 51 79

• light (practically white) b medium (light brown to dark brown) 0 dark (dark brown to black)

170 SOCIALIZATION OF THE CHILD

uine expression of actual preference, bu a reflection of ego involvement. Thi potential distortion of the data was con­ trolled by merely asking the children to indicate their preferences first and then to make identifications with one of the dolls.

t s

SUBJECTS

Two hundred fifty-three Negro chil dren formed the subjects of this experi ment. One hundred thirty-four of these subjects (southern group) were tested i segregated nursery schools and public schools in Hot Springs, Pine Bluff, and Little Rock, Arkansas. These children had had no experience in racially mixed school situations. One hundred nineteen subjects (northern group) were tested in the racially mixed nursery and public schools of Springfield, Massachusetts.

Age distribution of subjects :

­ ­

n

All subjects were tested individually in a schoolroom or office especially pro­ vided for this purpose. Except for a few children who showed generalized nega­ tivism from the beginning of the experi­ ment (results for these children are not included here), there was adequate rap­ port between the experimenter and all subjects tested. In general, the children showed high interest in and enthusiasm for the test materials and testing situ­ ation. The children, for the most part, considered the experiment somewhat of a game.

RESULTS

Racial Identification. Although the questions on knowledge of "racial dif­ ferences" and self-identification followed those designed to determine racial pref­ erence in the actual experimental situ­ ation, it appears more meaningful to discuss the results in the following order: knowledge of "racial differences," racial self-identification, and finally racial pref­ erences.

The results of the responses to requests 5, 6, and 7, which were asked to deter­ mine the subjects' knowledge of racial differences, may be seen in Table 1. Ninety-four percent of these children chose the white doll when asked to give the experimenter the white doll; 93 per­ cent of them chose the brown doll when asked to give the colored doll; and, 72 percent chose the brown doll when asked to give the Negro doll. These re­ sults indicate a clearly established knowl­ edge of a "racial difference" in these subjects-and some awareness of the relation between the physical character­ istic of skin color and the racial concepts of "white" and "colored." Knowledge of the concept of "Negro" is not so well developed as the more concrete verbal concepts of "white" and "colored" as applied to racial differences.

The question arises as to whether choice of the brown doll or of the white oll, particularly in response to ques-d

RACIAL IDENTIFICATION AND PREFERENCE 171

TABLE 1

CHOICES OF ALL SUBJECTS

Request 5 Request 6 Request 7 Request 8 (for white) (for colored) (for Negro) (for you) Choice

No. Percent —

No. Percent No. Percent No. – —– —

Percent – – – — – —

Colored doll 13 5 235 93 182 72 166 66 White doll. 237 94 15 6 50 20 85 33 Don't know or no response 3 1 .) 1 21 8 2 1

tions S and 6, really reveals a knowledge of "racial differences" or simply indi­ cates a learned perceptual reaction to the concepts of "colored" and "white." Our evidence that the responses of these children do indicate a knowledge of "racial difference" comes from several sources: the results from other tech­ niques used (i.e.', a coloring test and a questionnaire) and from the qualitative data obtained (children's spontaneous remarks) strongly support a knowledge of "racial differences." Moreover, the consistency of results for requests S through 8 also tends to support the fact that these children are actually making identifications in a "racial" sense.

The responses to request 8, designed to determine racial self-identification fol­ low the following pattern: 66 percent of the total group of children identified themselves with the colored doll, while 33 percent identified themselves with the white doll. The critical ratio of this difference is 7.6.2

Comparing the results of request 8 (racial self-identification) with those of requests 5, 6, and 7 (knowledge of racial difference) it is seen that the awareness of racial differences does not necessarily determine a socially accurate racial self­ identification-since approximately nine out of ten of these children are aware of racial differences as indicated by their correct choice of a "white" and "colored" doll on request, and only a

little more than six out of ten make socially correct identifications with the colored doll.

Age Differences. Table 2 shows that, when the responses to requests S and 6 are observed together, these subjects at each age level have a well-developed knowledge of the concept of racial differ­ ence between "white" and "colored" as this is indicated by the characteristic of skin color. These data definitely indicate that a basic knowledge of "racial differ­ ences" exists as a part of the pattern of ideas of Negro children from the age of three through seven years in the north­ ern and southern communities tested in this study- and that this knowledge de­ velops more definitely from year to year to the point of absolute stability at the age of seven.

A comparison of the results of re­ quests S and 6 with those of request 7, which required the child to indicate the doll which looks like a "Negro" child, shows that knowledge of a racial differ­ ence in terms of the word "Negro" docs not exist with the same degree of defi­ niteness as it does in terms of the more basic designations of "white" and "colored." It is significant, however, that knowledge of a difference in terms of the word "Negro" makes a sharp increase from the five- to the six-year level and a less accelerated one between the six­ and seven-year levels. The fact that all of the six-year-olds used in this investi-

2 These results are supported by similar ones from the Horowitz line drawing technique.

172 SOCIALIZATION OF THE CHILD

TABLE 2

CHOICES OF SUBJECTS AT EACH AGE LEVEL*

3 yr. 4 yr. 5 yr. 6 yr. 7 yr.

Choice Per- Per- Per- Per- Per-

No. No. No. No. No. cent cent cent cent cent

————- Request 5 (for white)

colored doll 4 13 4 14 3 7 2 3 0 white doll 24 77 25 86 43 94 70 97 75 100

Request 6 (for colored)

colored doll 24 77 24 83 43 94 69 96 75 100 white doll 4 13 5 17 3 7 3 4 0

Request 7 (for Negro)

colored doll 17 55 17 59 28 61 56 78 64 85 white doll 9 29 10 35 14 30 12 17 5 7

Request 8 (for you)

colored doll 11 36 19 66 22 48 49 68 65 87 white doll . 19 61 9 31 24 52 23 32 10 13

• Individuals failing to make either choice not included, hence some percentages add to less than 100.

gation were enrolled in the public schools seems to be related to this spurt. Since it seems clear that the term "Negro" is a more verbalized designa­ tion of "racial differences," it is reason­ able to assume that attendance at public schools facilitates the development of this verbalization of the race concept held by these children.

In response to request 8 there is a general and marked increase in the per­ cent of subjects who identify with the colored doll with an increase in age­ with the exception of the four- to five­ year groups.3 This deviation of the five­ year-olds from the general trend is con­ sidered in detail in the larger, yet un­ published study.

Identification by Skin Color. Table 3

shows slight and statistically insignificant differences among the three skin-color groups in their responses which indicate a knowledge of the "racial difference " between the white and colored doll (requests 5 through 7).

It should be noted, however, that the dark group is consistently more accurate in its choice of the appropriate doll than either the light or the medium group on requests 5 through 7. This would seem to indicate that the dark group is slightly more definite in its knowledge of racial differences and that this definiteness ex­ tends even to the higher level of verbali­ zation inherent in the use of the term "Negro" as a racial designation. In this regard it is seen that 75 percent of the dark children chose the colored doll

3 These results are supported by those from the use of the Horowitz line drawing technique.

RACIAL IDENTIFICATION AND PREFERENCE 173

TABLE 3

CHOICES OF SUBJECTS IN LICHT, MEDIUM, ANO DARK GROUPS*

Light Medium Dark Choice

No. Percent No. Percent No. Percent

Request 5 (for white)

colored doll 2 5 8 6 3 4 white doll 43 94 118 92 76 96

Request 6 (for colored)

colored doll 41 89 118 92 76 96 white doll 4 9 8 6 3 4

Request 7 (for Negro)

colored doll 32 70 91 71 59 75 white doll 9 20 27 21 14 18

Request 8 (for you)

colored doll 9 20 93 73 64 81 white doll 37 80 33 26 15 19

* Individuals failing to make either choice not included, hence some percentages add to less than 100.

when asked for the doll which "looks like a Negro child" while only 70 percent of the light children and 71 percent of the medium children made this response. The trend of results for requests 5 and 6 remains substantially the same.

These results suggest further that cor­ rect racial identification of these Negro children at these ages is to a large extent determined by the concrete fact of their own skin color, and further that this racial identification is not necessarily dependent upon the expressed knowledge of a racial difference as indicated by the correct use of the words "white," "colored," or "Negro" when responding to white and colored dolls. This conclu­ sion seems warranted in the light of the fact that those children who differed

in skin color from light through medium to dark were practically similar in the pattern of their responses which indi­ cated awareness of racial differences but differed markedly in their racial identi­ fication (responses to request 8 for the doll "that looks like you") only 20 per­ cent of the light children, while 73 per­ cent of the medium children, and 81 percent of the dark children identified themselves with the colored doll.

It is seen that there is a consistent in­ crease in choice of the colored doll from the light to the medium group; an in­ crease from the medium group to the dark group; and, a striking increase in the choices of the colored doll by the dark group as compared to the light group.4 All differences, except between

• These results substantiate and clearly focus the trend observed through the use of the Horowitz line drawing technique.

174 SOCIALIZATION OF THE CHILD

TABLE 4

CHOICES OF SUBJECTS IN NORTHERN (MIXED SCHOOLS) AND SOUTHERN

(SEGREGATED SCHOOLS) GROUPS*

North, South, Choice percent percent

Request 5 (for white) colored doll 4 6 white doll . 94 93

Request 6 (for colored) colored doll 92 94 white doll . 7 5

Request 7 (for Negro) colored doll 74 70 white doll . 20 19

Request 8 (for you) colored doll 61 69 white doll . 39 29

*Individuals failing to make either choice not included, hence some percentages add to le-.is than 100.

the medium and dark groups, are statis­ tically significant.

Again, as in previous work,5 it is shown that the percentage of the medium groups' identifications with the white or the colored representation resembles more that of the dark group and differs from the light group. Upon the basis of these results, therefore, one may assume that some of the factors and dynamics involved in racial identification are sub­ stantially the same for the dark and · medium children, in contrast to dynam­ ics for the light children.

North-South Differences. The results presented in Table 4 indicate that there are no significant quantitative differences between the northern and southern Negro children tested (children in mixed schools and children in segregated

schools) in their knowledge of racial differences.

While none of these differences is sta­ tistically reliable, it is significant that northern children know as well as southern children which doll is supposed to represent a white child and which doll is supposed to represent a colored child. However, the northern children make fewer identifications with the colored doll and more identifications with the white doll than do the southern children. One factor accounting for this difference may be the fact that in this sample there are many more light colored children in the North (33) than there are in the South (13). Since this difference in self-identifi­ cation is not statistically significant, it may be stated that the children in the northern mixed-school situation do not

6 K. B. and M . P. Clark, "Skin Color as a Factor in Racial Identification of Negro Preschool Children," J. Soc. Psychol., 1940, XI, 159–169; "Segregation as a Factor in the Racial Identification of Negro Preschool Chlldren: a preliminary report," J . &per. Educ., 1939, IX, 161-163; "The Development of Consciousness of Self and the Emergence of Racial Identification in Negro Pre­ school Children," J. Soc. Psychol., 1939, X, 591-599.

RACIAL IDENTIFICATION AND PREFERENCE 175

TABLE 5

Cuo1CEs OF ALL SUBJECTS

Request 1 Request 2 Reque:;t 3 Request 4 (play with) (nice doll) (looks bad) (nice color) Choice

No. Percent No. — Percent

– No. – –

Percent No. Percent – – – ——

Colored doll 83 32 97 38 149 59 96 38 White doll. 169 67 150 59 42 17 151 60 Don't know or

no response l l 6 3 62 24 6 2

differ from children in the southern segregated schools in either their knowl­ edge of racial differences or their racial identification. A more qualitative analy­ sis will be presented elsewhere.

Racial Preferences. It is clear from Table 5 that the majority of these Negro children prefer the white doll and reject the colored doll.

Approximately two thirds of the sub­ jects indicated by their responses to re­ quests 1 and 2 that they like the white doll "best," or that they would like to play with the white doll in preference to the colored doll, and that the white doll is a "nice doll."

Their responses to request 3 show that this preference for the white doll implies a concomitant negative attitude to­ ward the brown doll. Fifty-nine percent of these children indicated that the colored doll "looks bad," while only 17 percent stated that the white doll "looks bad" (critical ratio 10.9). That this preference and negation in some way involve skin color is indicated by the results for request 4. Only 38 percent of the children thought that the brown doll was a "nice color," while 60 percent of them thought that the white doll was a "nice color" (critical ratio 5.0).

The importance of these results for an understanding of the origin and develop­ ment of racial concepts and attitudes in Negro children cannot be minimized. Of equal significance are their implications, in the light of the results of racial identi-

:fication already presented, for racial mental hygiene.

Age Differences. Table 6 shows that at each age from three through seven years the majority of these children prefer the white doll and reject the brown doll. This tendency to prefer the white doll is not as stable (not statistically reliable) in the three-year-olds as it is in the four­ and five-year-olds. On the other hand, however, the tendency of the three-year­ olds to negate the brown doll (" looks bad") is established as a statistically significant fact (critical ratio 4.5).

Analyzing the results of requests 1 and 2 together, it is seen that there is a marked increase in preference for the white doll from the three- to the four-year level; a more gradual decrease in this preference from the four- to the five-year level; a further decrease from the five- to the six­ year level; and a continued decrease from the six- to the seven-year level. These results suggest that although the majority of Negro children at each age prefer the white doll to the brown do11, this preference decreases gradually from four through seven years.

Skin color preferences of these children follow a somewhat different pattern of development. The results of request 4 show that while the majority of children at each age below 7 years prefer the skin color of the white doll, this preference increases from three through five years and decreases from five through seven years. It is of interest to point out that

CHOICES OF SUBJECTS AT EACH AGE LEVEL*

3 yr. 4 yr. 5 yr. 6 yr. 7 yr.

Choice Per- Per- Per- Per- Per-

No. No. No. No. No. cent cent cent cent cent ——————

Request 1 (play with)

colored doll 13 42 7 24 12 26 21 29 30 40 white doll 17 55 22 76 34 74 51 71 45 60

Request 2 (nice doll)

colored doll 11 36 7 24 13 28 33 46 33 44 white doll 18 58 22 76 33 72 38 53 39 52

Request 3 (looks bad)

colored doll 21 68 15 52 36 78 45 63 32 43 white doll 6 19 7 24 5 11 11 15 13 17

Request 4 (nice color)

colored doll 12 39 8 28 9 20 31 43 36 48 white doll . 18 58 21 72 36 78 40 56 36 48

* Individuals failing to make either choice not included, hence some percentages add to less than 100.

TABLE 6

TABLE 7

CHOICES OF SUBJECTS IN LIGHT, MEDIUM, AND DARK GROUPS*

Light Medium Dark Choice

No. Percent No. Percent No. Percent

Request 1 (play with) colored doll 11 24 41 32 31 39 white doll 35 76 86 67 48 61

Request 2 (nice doll) colored doll 15 33 50 39 32 40 white doll 31 67 72 56 47 60

Request 3 (looks bad) colored doll 31 67 73 57 45 57 white doll 6 13 22 17 14 18

Request 4 (nice color) colored doll 13 28 56 44 27 34 white doll 32 70 68 53 51 65

* Individuals failing to make either choice not included, hence some percentages add to less than 100.

RACIAL IDENTIFICATION AND PREFERENCE

TABLE 8

CHOICES OF SUBJECTS IN NORTHERN (MIXED SCUOOLS) AND SounmRN (SEGREGATED SCHOOLS) GROUPS (REQUESTS 1 TiIROUGH 4)*

North, South, Choice percent percent

Request 1 (play with) colored doll 28 37 white doll . 72 62

Request 2 (nice doll) colored doll 30 46 white doll . 68 52

Request 3 (looks bad) colored doll 7l 49 white doll . 17 16

Request 4 (nice color) colored doll 37 40 white doll . 63 57

• Individuals miling to make either choice not included, hence some percentages add to less than 100.

177

only al the seven-year level do the same number of children indicate a preference for the skin color of the colored doll as for that of the white doll.

The majority of these children at each age level indicate that the brown doll, rather than the white doll, "looks bad." This result shows positively the nega­ tion of the colored doll which was im­ plicit in the expressed preference for the white doll discussed above.

The evaluative rejection of the brown doll is statistically significant, even at the three-year level, and is pronounced al the five-year level. The indicated pref­ erence for the white doll is statistically significant from the four-year level up to the seven-year level.

It seems justifiable to assume from these results that the crucial period in the formation and patterning of racial attitudes begins at around four and five years. At these ages these subjects appear to be reacting more uncritically in a definite structuring of attitudes which

conforms with the accepted racial value and mores of the larger environment.

Preferences and Skin Color. Result presented in Table 7 reveal that there i a tendency for the majority of these chil­ dren, in spite of their own skin color, t prefer the white doll and to negate th brown doll. This tendency is most pro nounced in the children of light skin color and least so in the dark children. A more intensive analysis of these results appearsinalarger,yet unpublished study.

North-South Differences. From Table 8 it is clear that the southern children in segregated schools are less pronounced in their preference for the white doll, compared to the northern children's defi­ nite preference for this doll. Although still in a minority, a higher percentage of southern children, compared to northern, prefer to play with the colored doll or think that it is a "nice" doll. The criti­ cal ratio of this difference is not signifi­ cant for request 1 but approaches sig­ nificance for request 2 (2.75).

s

s s

o e ­

178 SOCIALIZATION OF THE CHILD

A significantly higher percentage (71) of the northern children, compared to southern children (49) think that the brown doll looks bad (critical ratio 3.68). Also a slightJy higher percent of the southern children think that the brown doll has a" nice color," while more north­ ern children think that the white doll has a "nice color."

In general, it may be stated that north­ ern and sou them children in these age groups tend to be similar in the degree of their preference for the white doll­ with the northern children tending to be somewhat more favorable to the white doll than are the southern children. The southern children, however, in spite of their equal favorableness toward the white doll, are significantly less likely to reject the brown doll (evaluate it nega­ tively), as compared to the strong tend­ ency for the majority of the northern children to do so. That this difference is not primarily due to the larger number of light children found in the northern sample is indicated by more intensive analysis presented in the complete report.

Some Qualitative Data. Many of the children entered into the experimental situation with a freedom similar to that of play. They tended to verbalize freely and much of this unsolicited verbalization was relevant to the basic problems of this study.

On the whole, the rejection of the brown doll and the preference for the white doll, when explained at all, were explained in rather simple, concrete terms: for white-doll preference-" 'cause he's pretty" or " 'cause he's white"; for

rejection of the brown doll-" 'cause he's ugly" or "'cause it don't look pretty" or " 'cause him black" or "got black on him."

On the other hand, some of the chil­ dren who were free and relaxed in the beginning of the experiment broke down and cried or became somewhat nega­ tivistic during the latter part when they were required to make self-identifica­ tions. Indeed, two children ran out of the testing room, unconsolable, con­ vulsed in tears. Thi:. type of behavior, although not so extreme, was more prev­ alent in the North than in the South. The sou them children who were disturbed by this aspect of the experiment generally indicated their disturbance by smiling or matter of factly attempting to escape their dilemma either by attempted humor or rationalization.

Rationalization of the rejection of the brown doll was found among both north­ ern and southern children, however. A northern medium six-year-old justified his rejection of the brown doll by stating that "he looks bad 'cause he hasn't got a eyelash." A seven-year-old medium northern child justified his choice of the white doll as the doll with a" nice color" because "his feet, hands, ears, elbows, knees, and hair are clean."

A northern five-year-old dark child felt compelled to explain his identifica­ tion with the brown doll by making the following unsolicited statement: " I burned my face and made it spoil." A seven-year-old northern light child went to great pains to explain that he is ac­ tually white but: "I look brown because I got a suntan in the summer."

The"VisuaI Cliff"

This simple apparatus is used to investigate depth perception in different anitnals. All species thus far tested seem able to

perceive and a void a sharp drop as soon as they can move about

by Eleanor J. Gibson and Richard D. Walk

H uman infants at the creeping and

toddling stage are notoriously prone to falls from more or less

high places. They must be kept from going over the brink by side panels on their cribs, gates on stairways and the vigilance of adults. As their muscular coordination matures they begin to avoid such accidents on their own. Common sense might suggest that the child learns to recognize falling-off places by experi­ ence-that is, by falling and hurting him­ self. But is experience really the teacher? Or is the ability to perceive and avoid a brink part of the child's original endow­ ment?

Answers to these questions will throw light on the genesis of space perception in general. Height perception is a special case of distance perception: information in the light reaching the eye provides stimuli that can be utilized for the dis­ crimination both of depth and of reced­ ing distance on the level. At what stage of development can an animal respond effectively to these stimuli? Does the on­ ,et of such response vary with animals of different species and habitats?

At Cornell University we have been investigating these problems by means of a simple experimental setup that we call a visual cliff. The cliff is a simulated one and hence makes it possible not only to control the optical and other stimuli (auditory and tactual, for instance) but also to protect the experimental subjects. It consists of a board laid across a large sheet of heavy glass which is supported a foot or more above the Boor. On one side of the board a sheet of patterned material is placed Bush against the un­ dersurface of the glass, giving the glass the appearance as well as the substance of solidity. On the other side a sheet of the same material is laid upon the Boor; this side of the board thus becomes the

visual cliff [see photograph on cover}. We tested 36 infants ranging in age

from six months to 14 months on the visual cliff. Each child was placed upon the center board, and his mother called him to her from the cliff side and the shallow side successively. All of the 27 infants who moved off the board crawled out on the shallow side at least once; only three of them crept off the brink onto the glass suspended above the pat­ tern on the Boor. Many of the infants crawled away from the mother when she called to them from the cliff side; others cried when she stood there, because they could not come to her without crossing an apparent chasm. The experiment thus demonstrated that most human infants can discriminate depth as soon as they can crawl.

The behavior of the children in this situation gave clear evidence of their

dependence on vision. Often they would peer down through the glass on the deep side and then back away. Others would pat the glass with their hands, yet de­ spite this tactual assurance of solidity would refuse to cross. It was equally clear that their perception of depth had matured more rapidly than had their locomotor abilities. Many sup­ ported themselves on the glass over the deep side as they maneuvered awk­ wardly on the board; some even backed out onto the glass as they started toward the mother on the shallow side. Were it not for the glass some of the children would have fallen off the board. Evident­ ly infants should not be left close to a brink, no matter how well they may discriminate depth.

This experiment does not prove that the human infant's perception and avoidance of the cliff are innate. Such an interpretation is supported, however, by

the experiments with nonhuman infants. On the visual cliff we have observed the behavior of chicks, turtles, rats, lambs, kids, pigs, kittens and dogs. These ani­ mals showed various reactions, each of which proved to be characteristic of their species. In each case the reaction is plainly related to the role of vision in the survival of the species, and the varied patterns of behavior suggest something about the role of vision in evolution.

In the chick, for example, depth per­ ception manifests itself with special rapidity. At an age of less than 24 hours the chick can be tested on the visual cliff. It never makes a "mistake" and al­ ways hops off the board on the shallow side. Without doubt this finding is re­ lated to the fact that the chick, unlike many other young birds, must scratch for itself a few hours after it is hatched.

Kids and lambs, like chicks, can be tested on the visual cliff as soon as they can stand. The response of these animals is equally predictable. No goat or lamb ever stepped onto the glass of the deep side, even at one day of age. When one of these animals was placed upon the glass on the deep side, it displayed char­ acteristic stereotyped behavior. It would refuse to put its feet down and would back up into a posture of defense, its front legs rigid and its hind legs limp. In this state of immobility it could be pushed forward across the glass until its head and field of vision crossed the edge of the surrounding solid surface, where­ upon it would relax and spring forward upon the surface.

At the Cornell Behavior Farm a group of experimenters has carried these exper­ iments with kids and goats a step further. They fixed the patterned material to a sheet of plywood and were thus able to adjust the "depth" of the deep side. With the pattern held immediately be-

© 1960 SCIENTIFIC AMERICAN, INC

64

CHILD'S DEPTH PERCEPTION is tested on the visual cliff. The (top left), the child crawls to its mother across the "shallow" side apparatus consists of a board laid across a sheet of heavy glass, (top right). Called from the "deep" side, he pats the glass (bottom with a patterned material directly beneath the glass on one side left), but despite this tactual evidence that the "cliff" is in fact a and several feet below it on the other. Placed on the center board solid surface he refuses to cross over to the mother (bottom right) .

© 1960 SCIENTIFIC AMERICAN, INC

65

neath the glass, the animal would move to function without optical support. by smell, when moving about in the about the glass freely. With the optical Their sense of security or danger con­ dark, it responds to tactual cues from the Hoor dropped more than a foot below the tinued to depend upon the visual· cues stiff whiskers (vibrissae) on its snout. glass, the animal would immediately that give them their perception of depth. Hooded rats tested on the visual cliff freeze into its defensive posture. Despite The rat, in contrast, does not depend show little preference for the shallow repeated experience of the tactual solid­ predominantly upon visual cues. Its noc­ side so long as they can feel the glass ity of the glass, the animals never learned turnal habits lead it to seek food largely with their vibrissae. Placed upon the

KITTEN'S DEPTH PERCEPTION also manifests itself at an early age. Though the animal displays no alarm on the shallow side (top),

it "freezes" when placed on the glass over the deep side (bot· tom) ; in some cases it will crawl aimlessly backward in a circle.

© 1960 SCIENTIFIC AMERICAN, INC

66

glass over the deep side, they move about normallv. But when we raise the center board s�veral inches, so that the glass is out of reach of their whiskers, they evince good visual depth-discrimina­ tion: 95 to 100 per cent of them descend on the shallow side.

Cats, like rats, are nocturnal animals, sensitive to tactual cues from their

vibrissae. But the cat, as a predator, must rely more strongly on its sight. Kittens proved to have excellent depth-discrimi­ nation. At four weeks-about the earliest age that a kitten can move about with any facility-they invariably choose the shallow side of the cliff. On the glass over the deep side, they either freeze or circle aimlessly backward until they reach the center board [see illustratiol1s 011 oppo­ site page].

The animals that showed the poorest performance in our series were the tur­ tles. The late Robert M. Yerkes of Har­ vard University found in 1904 that aquatic turtles have somewhat poorer depth-discrimination than land turtles. On the visual cliff one might expect an aquatic turtle to respond to the reHec­ tions from the glass as it might to water and so prefer the deep side. They showed no such preference: 76 per cent of the aquatic turtles crawled off the board on the shallow side. The rela­ tively large minority that choose the deep side suggests either that this turtle has poorer depth-discrimination than other animals, or that its natural habitat gives it less occasion to "fear" a fall.

All of these observations square with what is known about the life history and ecological niche of each of the animals tested. The survival of a species re­ quires that its members develop dis­ crimination of depth by the time they take up independent locomotion, wheth­ er at one day (the chick and the goat), three to four weeks (the rat and the cat) or six to 10 months (the human infant) . That such a vital capacity does not de­ pend on possibly fatal accidents of learn­ ing in the lives of individuals is con­ sistent with evolutionary theory.

To make sure that no hidden bias was concealed in the design of the visual cliff we conducted a number of control experiments. In one of them we elimi­ nated reHections from the glass by light­ ing the patterned surfaces from below the glass (to accomplish this we dropped the pattern below the glass on both sides, but more on one side than on the other) . The animals-hooded rats-still consist­ ently chose the shallow side. As a test of the role of the patterned surface we

GOATS SHOW DEPTH PERCEPTION at an age of only one day. A kid walks freely on

the shallow side (top) ; on the deep side (middle) it leaps the "chasm" 10 safety (bo//,om).

© 1960 SCIENTIFIC AMERICAN, INC

67

– – – – – – – ——-;,. – – –

– – – , –

– – – • – –

• •

– – — >.

– – – – – –

., • – • • • – • – …. ,. .""

I

:

– – – –

– • –

– – –

.. .

• – • –

t

• : t

I

,, I

I

," I

TWO TYPES OF VISUAL DEPTH·CUE are diagrammed sche· in the fields. The spacing of the pattern elements (solid color)

matically on this page. Ellipses approximate the visual field of an decreases sharply heyond the edge of the cliff (tol1). The op·

animal standing near the edge of tbe cliff and looking toward it; tical motion (shaded color) of the elements as the animal moves

diagrams at right give the geometrical explanation of differences forward (center) or sideways (bottom) shows a similar drop·off.

© 1960 SCIENTIFIC AMERICAN, INC

68

–

– –

. –

. –

replaced it on either side of the cen­ terboard with a homogeneous grai sur­ face. Confronted with this choice, the rats showed no preference for either the shallow or the deep side. We also elimi­ nated the optical difference between the two sides of the board by placing the patterned surface directly against the undersurface of the glass on each side. The rats then descended without pref­ erence to either side. 'vVhen we lowered the pattern 10 inches below the glass on each side, they stayed on the board.

We set out next to determine which of two visual cues plays the de­

cisive role in depth perception. To an eye above the center board the optical pattern on the two sides differs in at least two important respects. On the deep side distance decreases the size and spacing of the pattern elements pro­ jected on the retina. "Motion parallax," on the other hand, causes the pattern elements on the shallow side to move more rapidly across the field of vision when the animal moves its position on the board or moves its head, just as nearby objects seen from a moving car appear to pass by more quickly than dis­ tant ones [see illustration on opposite page ]. To eliminate the potential dis­ tance cue provided by pattern density we increased the size and spacing of the pattern elements on the deep side in proportion to its distance from the eye [see top illustration at right]. With only the cue of motion parallax to guide them, adult rats still preferred the shallow side, though not so strongly as in the standard experiment. Infant rats chose the shallow side nearly 100 per cent of the time under both conditions, as did day-old chicks. Evidently both species can discriminate d'Jpth by dif­ ferential motion alone, with no aid from texture density and probably little help from other cues. The perception of dis­ tance by binocular parallax, which doubtless plays an important part in 11U­ man behavior, would not seem to have a significant role, for example, in the depth perception of chicks and rats.

To eliminate the cue of motion paral­ lax we placed the patterned material di­ rectly against the glass on either side of the board but used smaller and more densely spaced pattern-elements on the cliff side. Both young and adult hooded rats preferred the side with the larger pattern, which eviden�ly "signified" a nearer surface. Day-old chicks, however, showed no preference for the larger pat­ ter.n. It may be that learning plays some part in the preference exhibited by the

rats, since the young rats were tested at a somewhat older age than the chicks. This supposition is supported by the re­ sults of our experiments with animals reared in the dark.

The effects of early experience and of such deprivations as dark-rearing repre­ sent important clues to the relative roles of maturation and learning in animal be­ havior. The first experiments along this line were performed by K. S. Lashley and James T. Russell at the Univer­ sity of Chicago in 1934. They tested light-reared and dark-reared rats on a

"jumping stand" from which they in­ duced animals to leap toward a platform placed at varying distances. Upon find­ ing that both groups of animals jumped with a force closely correlated with dis­ tance, they concluded that depth per­ ception in rats is innate. Other investi-

gators have pointed out, however, that the dark-reared rats required a certain amount of "pretraining'; in the light be­ fore they could be made to jump. Since the visual-cliff technique requires no pretraining, we employed it to test groups of light-reared and dark-reared hooded rats. At the age of 90 days both groups showed the same preference for the shallow side of the apparatus, confirming Lashley's and Russell's con­ clusion.

Recalling our findings in the young rat, we then took up the question of whether the dark-reared rats relied upon motion parallax or upon contrast in tex­ ture density to discriminate depth. When the animals were confronted with the visual cliff, cued only by motion parallax, they preferred the shallow side, as had the light-reared animals. When the

SEPARATION OF VISUAL CUES is shown in these diagrams. Pattern density is held

constant (top) by using a larger pattern on the low side of the cliff; the drop in optical

motion (motion parallax) remains. Motion parallax is equalized (bottom) by placing patterns at same level; the smaller pattern on one side preserves difference in spacing.

© 1960 SCIENTIFIC AMERICAN, INC

69

•••

• •

• • •

•

IMPORTANCE OF PATTERN in depth perception is shown in left), presumably because it "signified" a nearer and therefore

these photographs. Of two patterns set at the same depth, normal safer surface. Confronted with two patternless surfaces set at dif.

rats almost invariahly preferred the larger (top row and bottom ferent depths, the animals displayed no preference (bottom right).

© 1960 SCIENTIFIC AMERICAN, INC

70

—r-'l—

0 —1.-1–­

�

�

�

choice was cued by pattern density, however, they departed from the pattern of the normal animals and showed no significant preference [see bottom illus­ tration at right]. The behavior of dark­ Teared rats thus resembles that of the day-old chicks, which also lack visual experience. It seems likely, therefore, that of the two cues only motion par­ allax is an innate cue for depth dis­ crimination. Responses to differential pattern-density may be learned later.

One cannot automatically extrapolate these results to other species. But

experiments with dark-reared kittens in­ dicate that in these animals, too, depth perception matures independently of trial and error learning. In the kitten, :however, light is necessary for normal visual maturation. Kittens reared in the dark to the age of 27 days at first crawled {)r fell off the center board equally often {)n the deep and shallow sides. Placed upon the glass over the deep side, they ·did not back in a circle like normal kit­ tens but showed the same behavior that they had exhibited on the shallow side. Other investigators have observed equiv­ alent behavior in dark-reared kittens; they bump into obstacles, lack nOlIDal eye movement and appear to "stare" .straight ahead. These difficulties pass after a few days in the light. We accord­ ingly tested the kittens every day. By the end of a week they were performing in every respect like normal kittens. They showed the same unanimous pref­ erence for the shallow side. Placed upon the glass over the deep side, they balked and circled backward to a visually se­ cure surface. Repeated descents to the deep side, and placement upon the glass during their "blind" period, had not taught them that the deep side was "safe." Instead they avoided it more and more consistently. The initial blindness of dark-reared kittens makes them ideal subjects for studying the mat­ uration of depth perception. With further study it should be possible to determine which cues they respond to first and what kinds of visual experi­ ence accelerate or retard the process of maturation.

From our first few years of work with the visual cliff we are re;ldy to venture the rather broad conclusion that a seeing animal will be able to discriminate depth when its locomotion is adequate, even when locomotion begins at birth. But many experiments remain to be done, especially on the role of different cues and on the effects of different kinds of early visual experience.

Vi UJ I u

I b: UJ Cl

< ::::> I-U «

2

4

8

ANIMALS DESCENDING (PER CENT)

CONTROL EXPERIMENT measured the effect on rats of reflections on the glass of the

apparatus. The percentage of animals leaving the center board decreased with increasing depth i.n much the same way, whether glass was present (black curve) or not ( colored curve) .

i='Z UJ U

a< UJ

<9 Z 0 Z UJ U V) UJ Cl

V) -' « Z«

100

80

60

40

20

RATS RATS KITTENS (MOTION (PATTERN (BOTH CUES)

PARALLAX CUES) DENSITY CUES)

DARK-REARING EXPERIMENTS reveal the order in which different depth-cues are utilized as animals mature. Animals reared in the light (open bars) all strongly preferred the shallow side (color) to the deep side (gray). Dark-reared'rats (solid bars), utilizing

motion parallax alone, still preferred the shallow side; pattern density alone elicited no preference. Dark-reared kittens also showed no preference, because of temporary blind­

ness. After seven days in the light aLI of them chose the shallow side (hatched bar)_

© 1960 SCIENTIFIC AMERICAN, INC

71

Child Development, May/June 2015, Volume 86, Number 3, Pages 800–811

Infants Learn Baby Signs From Video

Shoshana Dayanim and Laura L. Namy Emory University

There is little evidence that infants learn from infant-oriented educational videos and television programming. This 4-week longitudinal experiment investigated 15-month-olds’ (N = 92) ability to learn American Sign Lan- guage signs (e.g., patting head for hat) from at-home viewing of instructional video, either with or without parent support, compared to traditional parent instruction and a no-exposure control condition. Forced-choice, elicited production, and parent report measures indicate learning across all three exposure conditions, with a trend toward more robust learning in the parent support conditions, regardless of medium. There were no differences between experimental and control conditions in the acquisition of corresponding verbal labels. This constitutes the first experimental evidence of infants’ cially available educational videos.

High-quality educational television programming can be an effective source of learning in preschool- and school-aged children (e.g., Anderson, Huston, Schmitt, Linebarger, & Wright, 2001; Naigles & Mayeux, 2001; Rice, Huston, Truglio, & Wright, 1990). However, evidence supporting infants’ learn- ing from purportedly educational videos is more equivocal. Several studies suggest a negative corre- lation between overall media exposure and mea- sures of communicative development in infants under the age of 2 (Chonchaiya & Pruksananonda, 2008; Linebarger & Walker, 2005; Zimmerman, Christakis, & Meltzoff, 2007a), leading the Ameri- can Academy of Pediatrics [AAP] (2011) to renew its original recommendation (AAP, 1999) to avoid exposing children under 2 to television. Despite this recommendation, an estimated 90% of parents show television and videos to their infants (Zimm- erman, Christakis, & Meltzoff, 2007b) including many “educational” videos targeting infants. Vid- eos purporting to promote infants’ vocabulary and communicative development, in particular, have saturated the market (Vaala et al., 2010).

Shoshana Dayanim is now at the Department of Psychology, Keiser University.

We thank Diana Neeves, Saryn Levy, Cassie Hendrix, Kathleen Ghio, Trisha Patel, Evelyn Gilstein, and Jhonelle Bailey for their assistance with this study. We also thank the parents and partici- pants for their extraordinary dedication of time and effort to this research endeavor. This research was supported by NICHD Grant 5-R03-HD058777.

Correspondence concerning this article should be addressed to Shoshana Dayanim, Department of Graduate Psychology, Keiser University Graduate School, c/o 1380 Jody Lane, Atlanta, GA 30329. Electronic mail may be sent to sdayanim@keiseruniversi ty.edu.

ability to learn expressive communication from commer-

Few studies support the notion that educational videos can facilitate language acquisition in infancy. There is limited evidence that repeated exposure via video augments infants’ acquisition of words also heard in routine input, relative to infants who encounter the words in routine input alone (Lemish & Rice, 1986; Vandewater, 2011). Vandewater, Barr, Park, and Lee (2010) have also found that repeat- edly pairing words and shapes over a period of 15 days enables toddlers as young as 18 months to identify which shapes correspond to particular words.

There is, however, strong evidence that even when infants learn from video, there is a “video deficit” (Anderson & Pempek, 2005; Schmitt & Anderson, 2002)—attenuated learning relative to learning from live, interactive instruction (e.g., Barr & Hayne, 1999; Krcmar, Grela, & Lin, 2007). Fur- thermore, numerous studies report a failure to exhi- bit heightened learning of words introduced through a video medium in children younger than 24 months (see DeLoache et al., 2010; Krcmar, 2011; Robb, Richert, & Wartella, 2009).

This debate also extends to whether parental involvement in the viewing experience enhances learning. There is compelling evidence that parent coviewing enhances learning from television in pre- schoolers (e.g., Reiser, Williamson, & Suzuki, 1988; Roseberry, Hirsh-Pasek, Parish-Morris, & Golinkoff, 2009; Singer & Singer, 1998) and increases attention and verbal interactions during viewing in infants

© 2015 The Authors Child Development © 2015 Society for Research in Child Development, Inc. All rights reserved. 0009-3920/2015/8603-0010 DOI: 10.1111/cdev.12340

Infants’ Learning From Videos 801

(Barr, Zack, Muentener, & Garcia, 2008; Fender, Richert, Robb, & Wartella, 2010; Fidler, Zack, & Barr, 2010; Lemish, 1987). These consequences of parent coviewing may enhance the depth of infants’ cognitive processing of the video stimuli (Strouse, O’Doherty, & Troseth, 2013; Strouse & Troseth, 2014). However, recent longitudinal experimental studies (DeLoache et al., 2010; Robb et al., 2009) reported no evidence of heightened word learning in an infant coviewing condition, relative to view- ing alone or controls. Despite this lack of evidence that parent coviewing augments learning from tele- vision in infants, the AAP (2011) revised its guide- lines to suggest that if infant viewing is inevitable, coviewing is better than allowing an infant to view alone.

Although previous longitudinal studies of infants’ learning from videos have reported null effects, those studies focused exclusively on the acquisition of high-frequency words to which infants are exposed regularly outside of the video- viewing environment. Because exposure to the target stimuli was not fully controlled, there was evidence of word learning across conditions, includ- ing control conditions (e.g., DeLoache et al., 2010; Robb et al., 2009).

In the current longitudinal experiment, we asked whether better controlled exposure delivered exclu- sively through video would lead to any evidence of learning relative to a no-exposure control condition, and if so whether there was evidence of a video deficit, a benefit for parental coviewing, or both. To ensure experimental control of exposure, we investi- gated infants’ ability to learn symbolic gestures, or baby signs instead of focusing on word learning. Baby signs also offer better experimental control of exposure because they are not routinely employed in everyday interactions with infants. Thus, by using signs in place of spoken words, we are able to assess the independent impact of video exposure on learning more directly. Baby signs are a strong test case because infants begin using words and signs at around the same time, shortly after their first birthdays (see Acredolo & Goodwyn, 1988) and appear to use them for the same communica- tive purposes (Namy, 2001; Namy & Waxman, 1998). The availability of numerous educational vid- eos marketed toward infants that offer sign instruc- tion enables us to systematically and ecologically investigate the impact of medium (video vs. tradi- tional instruction) and parental involvement (co- viewing vs. alone) on infant learning.

We investigated 15-month-olds’ learning of baby signs from at-home viewing of commercially

available videos over the course of 3 weeks of exposure and also tested retention following 1 week without exposure. Acquisition of baby signs when viewing the videos alone or coviewing with parents was compared to traditional parent instruction and to a no-exposure control condition.

Method

Participants

Ninety-two 15-month-olds (M = 15.17 months at study onset, range = 13.95–16.81, SD = 7.04; 51 males) were recruited from the greater Atlanta area. The sample included approximately 74% Caucasian, 18% African American, 1% Asian, 1% Native Hawaiian or Other Pacific Islander, and 6% Mixed Ethnicities, with 8% identifying as being of His- panic or Latino descent.

Inclusion criteria included exposure to videos or screen media prior to recruitment contact, and lack of prior exposure to “baby signs.” Previous expo- sure to screen media was required to avoid induc- ing parents who were not already doing so to violate the AAP’s recommendation. No parents con- tacted were excluded for this criterion. Parents’ informed consent included acknowledgment of the AAP’s recommendation to avoid television expo- sure for infants under the age of 2.

Stimuli

We identified 18 target signs to use in the experi- ment. We selected only target signs that were object names to accommodate the use of still photographs of referents in learning assessments, and to mirror the types of labels (both verbal and gestural) most frequently acquired at this age. Target signs were selected based on familiarity of their referents as indexed by age of comprehension of their verbal labels (comprehended at a threshold of 50% of infants by M = 12.3 months; Fenson et al., 1994). These items included: airplane, apple, baby, ball, banana, bear, bird, book, car, cat, cookie, cracker, dog, fish, flower, hat, juice, and shoe. All referents were depicted in at least three different commer- cially available video productions intended to teach baby sign to infants.

Participants assigned to video-viewing conditions received a DVD compilation derived from six com- mercially available videos intended to teach infants baby signs. Each compilation included three chap- ters approximately 20 min in length with footage sampled from three to five videos in each chapter.

802 Dayanim and Namy

Each chapter included clips depicting the signs (and accompanying verbal labels) for each of the 18 objects. Parents were directed to rotate through the chapters across viewings to vary the order in which exposure to the signs occurred. We have opted not to identify the titles of the videos, as we did not seek to test the efficacy of individual video produc- tions. We were interested in assessing the impact of exposure to varying range of commercially avail- able videos. By declining to identify the video pro- ductions employed, we avoid any potential opportunities for the products’ marketers to make claims that may or may not be warranted based on this composite evidence.

There was variability in the formal features employed across the videos. These included (a) how each sign was introduced (e.g., introduction of sign occurs while an image of the referent was present vs. sequential presentation of the sign and the referent), (b) whether foreground or background music was employed, (c) whether still or moving images of the referents were presented, (d) how many scene changes were involved for each sign (ranging from 6 to approximately 21), and (e) the number of times the sign was repeated (ranging from approximately 3 to 15).

Although these formal features varied across pro- ductions, there were also several common factors across all sampled videos. Each video introduced signs in the first or second scene by showing an engaging person (either adult or child) producing a sign while simultaneously speaking the English ver- bal label for the referent. This introduction of the sign was consistently followed by several images of the referent, followed by a repetition of the sign and verbal label. For example, the video might depict an adult producing a sign for “dog” while saying “dog.” This would be followed by images of various types of dogs (real and toy dogs, still and moving pictures). Each clip concluded with the adult or child once again saying and signing “dog.”

Parents in the parent instruction condition intro- duced their infants to the 18 signs using a labora- tory-designed picture book in lieu of a video. Each page in the book was dedicated to one target sign (for a total of 18 pages) and included three different still photographs for each target. The stills were taken directly from screen shots included on the videos. A thumbnail picture of an adult producing the target sign with arrows signaling directions of motion was included in the top corner of each page to remind parents how to produce the sign. Appen- dix S1 in the online Supporting Information depicts sample pages from the book.

Each parent in the parental instruction condition also received a set of printed instructional materials on sign production including still frames from the sign videos demonstrating the sign being per- formed and pictures of the referents being labeled. These still photographs were augmented by sepa- rate printed verbal instructions describing how to perform each sign. The experimenter also demon- strated the signs to the parents in person during their baseline visit to the laboratory.

Procedure

Infants were randomly assigned to one of four conditions: video viewing alone (n = 20), video co- viewing with a parent (n = 27), a parent instruction condition that involved teaching signs using a pic- ture book (n = 21), and a no-sign-exposure control condition (n = 24). Seven additional infants who dropped out after 1 or 2 weeks were excluded from the study analysis (two each in video alone, parent instruction, and control group; one in the video co- viewing group). In all three experimental groups, parents were instructed to expose infants to 15– 20 min of sign instruction at home 4 days a week for 3 weeks, with no exposure to signs between instructional sessions. Given that the children tend to acquire the verbal labels for the 18 objects early in development, we expected children in all condi- tions to exhibit learning of the verbal labels for these objects due to routine, incidental exposure. As a result, we did not necessarily expect condition effects in word learning, despite the enhanced expo- sure to both words and signs for these referents resulting from participation in the study.

Video Viewing Groups

Parents in the video-alone condition were instructed not to interact with their child during viewing sessions. Those in the coviewing condition were instructed to watch with their child as they typically would at home and were told they could engage in any of the following behaviors: directing their child’s attention to the screen, imitating signs, and eliciting sign production from the child during viewing. Although such mediated (i.e., interactive and responsive) coviewing was encouraged, it was not mandated.

Parent Instruction Group

Parents in the parent instruction condition were instructed to introduce their infants to the 18 signs

Infants’ Learning From Videos 803

using the picture book. Parents were encouraged to teach their infants signs as they might teach new words from picture books at home and to point to the photographs and use verbal labels as well as signs. They were asked to limit instruction time to a maximum of 20 min per day, 4 days a week, to match exposure in the video viewing conditions.

Parents in all conditions were instructed to avoid using or imitating signs outside of the viewing or instructional sessions. If infants signed between instructional sessions, parents were asked to acknowledge the sign verbally and not to imitate it themselves. No instructions were specified regard- ing the use of verbal labels for the target objects outside of the viewing environment. Parents were asked to complete a diary at home documenting the date of each instructional session and noting if they saw their child producing a sign during a session or using a sign appropriately in between sessions. After completing 3 weeks of sign instruc- tion, parents were directed not to expose their infants to any signs for 1 week prior to returning to the laboratory at the end of the 4th week.

Learning Assessments

Children and parents in all conditions visited the laboratory weekly for 4 weeks for a total of five vis- its including baseline intake, to complete sign learn- ing assessments. Learning assessments included a weekly parent report checklist, a weekly forced- choice comprehension task, and a single elicited production task administered at the fifth visit after a week without at-home exposure.

Parent Report

Parents indicated weekly on a vocabulary check- list whether their infants comprehended or pro- duced appropriately each of the 18 target signs. They also indicated whether their infants compre- hended or produced the verbal labels correspond- ing to the signs (see Appendix S2 in the online Supporting Information).

Forced Choice

The laboratory-based forced-choice task was designed to measure sign comprehension. The experimenter, who sat across a table from the infant, placed photographs of two objects from the stimulus set (e.g., airplane and dog) on the table and produced the sign for one of them (e.g., the air- plane sign), asking the child, “Can you get it?” To

minimize fatigue, the experimenter administered six trials at each laboratory visit with the target items randomly selected. Across sessions, all items were tested at least once. Objects in each picture pair were matched for salience based on pilot testing with 15-month-olds (n = 19) who did not partici- pate in the experiment proper. During piloting, we selected pairings we believed were well matched for salience and presented them to the infants, ask- ing them to “get one.” Any pairings that elicited a bias toward selecting one object were altered by pairing more salient objects from one pair with more salient objects from another pair and readmin- istering the choice elicitation with additional infants to ensure that infants did not exhibit default sys- tematic preferences within any pairing.

Experimenters were generally blind to condition assignment, although conversations with parents occasionally inadvertently revealed to which condi- tion the infant was assigned. The experimenters were instructed to ensure that the two picture cards and sign production were equidistant from the child, and to ensure that their eye gaze while elicit- ing a choice remained fixed on the infant’s face. Coders were also blind to condition.

Elicited Production

On their final visit, in addition to completing the checklist and forced-choice comprehension task, infants in the experimental conditions also com- pleted an elicited sign production task as a conser- vative test of learning and retention. We also administered the elicited production task to five infants in the no-exposure control group but discon- tinued this with subsequent control participants due to the distress and confusion displayed by these infants. None of the five who participated produced any signs.

This measure was somewhat exploratory as infants of this age often fail to produce communica- tive signals in laboratory-based elicited production tasks. As a result, we expected that production would be low, but nonetheless had the potential to provide a compelling index of depth of learning. To elicit production, the experimenter presented photo- graphs of the target objects one at a time and elic- ited the sign by asking the infant, “Can you say [points to photograph] this with your hands?” or “Can you show me [verbal label] with your hands?” The number of trials administered varied across infants based on their attention and fussiness (M = 7.61 trials, SD = 3.92, range = 2–18). Eleven infants from the experimental groups did not com-

804 Dayanim and Namy

plete this task due to fussiness (4 = video alone, 3 = supported video, 4 = parent instruction). This resulted in a total sample size for this measure of 57 across the three experimental conditions.

Coding

A coder blind to condition classified the infants’ choices during the forced-choice task and the infants’ responses to the sign elicitation from video. Choice response was based on which card the infant first touched. Elicitation was coded by credit- ing infants with correct sign production if they pro- duced an intentional hand movement that included at least two of the three central elements of the sign: hand shape, motion trajectory, and sign-space loca- tion, which were each coded independently. A sec- ond coder evaluated a randomly selected 10% of the sessions. Intercoder agreement was 97% on forced-choice trials and 87% on elicited production.

Results

Next, we report the parent report data for both sign learning and word learning followed by the more conservative laboratory-based forced-choice sign comprehension and elicited sign production mea- sures. Data collected during the first 3 weeks of the study (four laboratory visits including baseline) were used to evaluate learning whereas data col- lected after the no-exposure delay were analyzed separately as a measure of retention.

Parent Report

We investigated infants’ baby sign learning as well as parents’ report of children’s use of verbal labels for each of the 18 target objects. We tracked verbal label acquisition both as a replication of previ- ous longitudinal research and as a manipulation check to ensure that there were no systematic report- ing biases exhibited by parents in particular conditions. For both sign and word production, we calculated the proportion of target items (of 18) that parents reported their children produced each week. We investigated whether parents reported growth in target sign and verbal label acquisition over time and whether this varied as a function of expo- sure condition using two-way analyses of variance (ANOVAs) with exposure condition (video alone, coviewing, parent instruction, and no-exposure con- trol) as a between-subjects variable and laboratory visit (baseline, 1, 2, 3) as a within-subject variable.

Sign Learning

We analyzed the proportion of signs parents reported infants produced using a two-way (Condi- tion 9 Laboratory Visit) ANOVA. This analysis yielded a main effect of condition, F(3, 88) = 7.01, p < .0005, partial g 2 = .19, and a main effect of lab- oratory visit, Λ = .49, F(3, 86) = 29.34, p < .0005, partial g 2 = .51, mediated by a Condition 9 Labo- ratory Visit interaction, Λ = .66, F(9, 88) = 4.33, p < .0005, partial g 2 = .13 (see Figure 1). This inter- action was driven by the control group, which devi- ated from the three experimental conditions. Whereas parents of infants in all three experimental groups reported growth in their children’s sign pro- duction across laboratory visits, those in the control condition did not. A follow-up analysis that excluded the control condition revealed only a main effect of laboratory visit, Λ = .43, F(3, 63) = 28.14, p < .0005, partial g 2 = .57, suggesting that the expo- sure conditions did not differ reliably from each other. Follow-up one-way ANOVAs comparing the four conditions for each laboratory visit indepen- dently revealed no condition effects at baseline or after 1 week of exposure, but a reliable condition effect after 2 weeks, F(3, 88) = 6.04, p = .001, and 3 weeks of exposure, F(3, 88) = 11.05, p < .0005, suggesting that parents in the experimental condi- tions began to observe evidence of learning after 2 weeks of exposure. Post hoc analysis using both Tamhane (to adjust for violation of homogeneity given that parents never reported any sign produc- tion in the control group) and Tukey’s honestly sig- nificant difference (HSD) test indicated that all three experimental conditions differed from control after 2 weeks’ exposure. None of the experimental conditions differed reliably from each other at any laboratory visit, although the difference between co- viewing and video alone approached significance at Laboratory Visit 3 (Tukey’s p = .064).

Word Learning

We predicted that acquisition of verbal labels for the included objects would increase over time at this age in all four conditions, due to incidental daily exposure. However, exposure to the baby signs also enhanced exposure to the accompanying verbal labels. To assess whether heightened exposure to the verbal labels in the sign exposure conditions acceler- ated word learning, we conducted an ANOVA on parent report of children’s word production for these 18 items with condition as a between-subject variable and laboratory visit (baseline, 1, 2, 3) as a

Infants’ Learning From Videos 805

within-subjects variable. As expected, there was a main effect of laboratory visit indicating vocabulary growth over time, Λ = .47, F(3, 86) = 31.67, p < .0005, partial g 2 = .52. However, there was no effect of condition and no interaction (see Figure 2).

Sign Retention

We conducted a one-way ANOVA to assess con- dition differences in sign retention as indicated by parental report at Laboratory Visit 4 (following the

1-week retention interval). This analysis yielded a significant condition effect, F(3, 87) = 14.68, p < .0005. Post hoc analysis using Tamhane revealed significant differences between the experi- mental groups and the control group (m = .0116), p < .01, but no differences among experimental groups. Tukey’s HSD, a less conservative measure, indicated that parents of infants in the coviewing condition (m = .442) group reported that their infants produced significantly more of the signs than those in the video-alone condition (m = .242),

0.50

Pr op

or on

S ig

ns P

ro du

ce d 0.40

0.30

0.20

0.10

0.00

Video Alone Co-Viewing Parent Instruc on No Exposure

Baseline Week 1 Week 2 Week 3

Figure 1. Mean proportion of signs accumulated across sessions in each condition, based on parental report. Error bars indicate confi- dence intervals (95%).

0.50

Pr op

or on

W or

ds P

ro du

ce d

0.40

0.30

0.20

0.10

0.00 Baseline Week 1 Week 2 Week 3

Video Alone Co-Viewing Parent Instruc on No Exposure

Figure 2. Mean proportion of words accumulated across sessions in each condition, based on parental report. Error bars indicate confi- dence interval (95%).

806 Dayanim and Namy

p = .025. Parents reported an intermediate level of sign production in the parent instruction condition (m = .310) that did not differ reliably from either of the other sign exposure conditions.

Forced-Choice Assessment

The proportion of forced-choice trials (of six) on which infants selected the target picture was calculated for each laboratory visit. Given the longi- tudinal nature of the study, there were occasional missing data points due either to a missed labora- tory visit or infant fussiness. Overall, there were a total of 21 missing data points of 460 planned labo- ratory visits (< 5%). These missing laboratory visits were distributed across 19 participants (1 in the video alone, 12 in the coviewing, 4 in the parent instruction, and 4 in the control condition). Missing data points were replaced by the mean of the remaining children in the same condition for the same laboratory visit (see Parent, 2013; Rubin, Wit- kiewitz, St. Andre, & Riley, 2007, for support for this approach).

Sign Learning

We conducted two sets of analyses on the labo- ratory-based forced-choice sign learning task. We compared performance in the experimental groups both to performance in the control condition and to chance (random, 50%) responding. We compared the performance of each condition (including the

0.75

0.70

0.65

0.60

0.55

0.50

0.45

0.40

0.35

0.30

Video Alone Co-Viewing

Pr op

or �

on T

ar ge

t Se

le ct

ed

control condition) to chance using single-sample t tests. Comparisons to the control condition (and among experimental conditions) involved two-way Condition 9 Laboratory Visit ANOVAs.

Comparisons to chance indicated, as expected, that control performance did not differ from chance at any laboratory visit. Children in the video-alone condition responded at chance at the baseline visit and after 1 and 2 weeks of exposure but performed at above chance rates after 3 weeks of viewing, t(19) = 4.15, p = .001. In the coviewing condition, the same pattern was observed with performance above chance only at Laboratory Visit 3, t(26) = 2.51, p = .018. The parent instruction group exhib- ited chance performance at baseline and Laboratory Visits 1 through 3, performing marginally above chance in Laboratory Visit 3, t(20) = 2.05, p = .053.

An ANOVA with condition as a between-subject factor and laboratory visit (baseline, 1, 2, 3) as a within-subject factor yielded a main effect of labora- tory visit, Λ = .90, F(3, 86) = 3.01, p = .035, partial g 2 = .09, indicating overall improvement with expo- sure. There was also a marginal effect of condition, F(3, 88) = 2.17, p = .097, partial g 2 = .07 (see Figure 3). The interaction was not significant, Λ = .89, F(9, 88) = 1.13, p = .342, partial g 2 = .04. Post hoc analysis using Tukey’s HSD indicated that the marginal main effect of condition was driven by a reliable overall difference between the video-alone scores (M = .55, SD = .16) and the control condition scores (M = .49, SD = .18), p = .049. No other pair- wise condition differences were significant.

Parent Instruc�on No Exposure

Baseline Week 1 Week 2 Week 3

Figure 3. Mean proportion of target selected across sessions in each condition, based on forced-choice task. Error bars indicate confi- dence interval (95%). Chance = .50.

Infants’ Learning From Videos 807

Comparing Parent Report and Forced-Choice Measures

Parent report of infants’ signing and performance in the forced-choice task at Laboratory Visit 3 were marginally significantly correlated collapsed across all conditions, r(84) = .178, p = .10. However, this correlation was nonsignificant when infants in the control condition were removed from the analysis, r(61) = .049, p = .70.

Sign Retention

We conducted a separate analysis of performance on the forced-choice task at Laboratory Visit 4 after a 1-week delay during which children were not exposed to the baby signs. Comparisons to chance indicated that only those in the parent instruction group performed significantly above chance after a 1-week retention interval, t(19) = 2.85, p = .010. A one-way ANOVA revealed no reliable differences across conditions at this laboratory visit (see Fig- ure 4a).

At the sign retention session, there was a signifi- cant correlation between parent report and perfor- mance on the forced-choice task collapsed across all conditions, r(84) = .247, p = .022. The correlation remained marginally significant when infants in the control condition were removed from the analysis, r(62) = .202, p = .11.

Elicited Production as a Measure of Sign Retention

Performance on the elicited production task administered at the final Laboratory Visit (after a 1-week delay with no exposure) was measured based on the proportion of signs elicited that infants produced. Children’s mean production across conditions is reported in Figure 4b. Because only five infants completed the task in the no-expo- sure control condition and none of those infants produced any signs, we did not analyze the data from this condition further. Single-sample t tests comparing each experimental condition to the expected population mean of zero (representing no knowledge of the target signs) indicated that infants in all three experimental groups showed significant evidence of learning (ts = 5.47, 5.94, and 6.93 for the video-alone, coviewing, and parent instruction conditions, respectively, all ps < .0005). A one-way ANOVA with condition (video alone, coviewing, and parent instruction) as a between-subjects vari- able revealed no significant differences in rates of elicited sign production across the sign exposure conditions, F(2, 52) = .18, p = .832.

Among those infants who participated in the elicited production task, 82% of video-alone infants (n = 16) and 100% of those in the coviewing (n = 23) and parental instruction groups (n = 17) produced at least one sign successfully. However, performance in the elicited production task was not correlated with either parent report, r(53) = �.039, p = .778, or forced-choice performance, r(52) = .004, p = .978, at Laboratory Visit 4 (collapsed across experimental conditions), suggesting that the pro- duction task may better serve as an existence proof for learning than an accurate index of how many signs were retained.

Discussion

These data indicate that infants under the age of 2 can learn baby signs from video, even without the support of parents during viewing. Parent report and laboratory-based assessments revealed striking evidence of learning after 3 weeks of exposure (12 viewings) in all three experimental conditions. The evidence for sign retention was mixed. The forced- choice measure suggested that only those in the parent instruction condition retained the signs after a week without exposure. However, infants in all three experimental conditions reliably produced signs in the laboratory after a 1-week delay. Because production is typically considered the more conservative measure, and evidence of production clearly implies comprehension, it appears that infants in all three exposure conditions retained at least some sign knowledge over a delay. However, given that the production measure relies on perfor- mative factors such as fatigue and shyness, it is not likely to be the most sensitive index of variability in retention across conditions. The absence of correla- tions between elicited production and either the parent report or forced-choice assessment under- scores that the production measure was a less sensi- tive index.

Interestingly, the more sensitive forced-choice data suggest that children in the parental instruc- tion condition exhibited a “sleeper” effect, exhibit- ing marginally above-chance performance in the forced-choice task after 3 weeks’ exposure but robust evidence of learning after an additional week without exposure. The superior performance of the parent instruction condition relative to the video-viewing conditions after a delay may reflect a video deficit in retention after as little as 1 week. Taken together, the findings from this study sug- gest the potential for video-based learning but hint

808 Dayanim and Namy

(a)

1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00

(b)

1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00

Video Alone Co-Viewing Parent Instruc�on No Exposure

* *

*

Video Alone Co-Viewing Parent Instruc�on No Exposure

Figure 4. (a) Mean proportion of target selected after a 1-week delay, based on forced-choice task. Error bars indicate confidence inter- val (95%). Chance = .50. (b) Mean proportion of elicited signs produced after a 1-week delay. Error bars indicate standard error. *Indi- cates reliable difference from zero, indicating learning (p < .001).

that the most robust retention occurs following tra- ditional parent instruction.

The parental report data echoed the laboratory- based evidence suggesting learning across all expo- sure conditions. According to parent report, all three sign exposure groups exhibited learning after 3 weeks’ exposure. The two parent-supported groups (i.e., coviewing and parental instruction) reported numerically, but not statistically, higher rates of learning than those who viewed videos alone, according to parental report. This evidence of more robust learning in parentally supported learning environments was predicted and may imply that the medium through which information

is presented is less critical to infant learning than the involvement of a parent in the learning endeavor. However, the fact that these differences emerged only in the parental report measure raises the possibility that this outcome may be due, at least in part, to either greater sensitivity to sign pro- duction or over-reporting among parents who had participated in the learning sessions.

Although we anticipated that the infants in the parent instruction condition would demonstrate evidence of learning, it is worth noting that parents in this condition reported that they found the instructional sessions challenging. The task may not have been especially naturalistic, given that parents

Infants’ Learning From Videos 809

needed to simultaneously manipulate the book, sus- tain their infants’ attention, and track which signs to use. That the parents were teaching material with which they were not especially familiar may also have limited their consistency or comfort with producing the signs during interactions with their infants. In contrast, the adults on the commercial videos were adept signers. This, in conjunction with the more dynamic information included in the vid- eos, may actually have limited learning potential in the parent instruction condition (Simcock, Garrity, & Barr, 2011).

It is notable that infants whose parents sup- ported learning were successful at acquiring signs from both picture books and videos. However, our most surprising finding is that those in the video- alone condition learned the signs as well. There are several factors that may have contributed to learn- ing in this condition. The first is that infants were exposed to the videos repeatedly over the course of the study and repetition seems to support learning from video (Barr, Muentener, & Garcia, 2007; Barr, Muentener, Garcia, Fujimoto, & Ch�avez, 2007; Strouse & Troseth, 2008). In addition, although signs serve the same communicative functions as words, they are based on manual movement; Numerous imitation studies demonstrate infants’ ability to learn a sequence of movement from a screen (Barr & Hayne, 1999; Barr, Muentener, & Garcia, 2007; Barr, Muentener, Garcia, Fujimoto, et al., 2007; Barr, Shuck, Salerno, Atkinson, & Linebarger, 2010; Barr & Wyss, 2008; Hayne, Her- bert, & Simcock, 2003; Meltzoff, 1988a, 1988b; Strouse & Troseth, 2008). Furthermore, verbal labels, included in the elicited production task, may have served as reminders for the movements (Barr & Wyss, 2008; Hayne & Herbert, 2004; Khu, Gra- ham, & Ganea, 2013). Although signs did serve as labels in this context, it may be that this manual form of labeling is easier to learn from the screen than is verbal labeling. Furthermore, the familiarity with the words and objects may have scaffolded learning by drawing infants’ attention to the move- ment associated with the familiar word and referent (Strouse & Troseth, 2014). The fact that viewing alone did not require infants to divide their visual attention between the screen and the parent might also have aided learning in this context (Strouse & Troseth, 2014).

Although not a direct goal of our study, we also tracked word learning across laboratory visits via parent report. As we anticipated, word learn- ing increased over time for the stimulus items involved in this study, but the rate was consistent

across all conditions, including the no-exposure control condition. This outcome replicates previous longitudinal studies suggesting no enhanced word learning following 4 weeks of exposure to a com- mercial video as measured by parent report (see Robb et al., 2009) and forced-choice assessments (see DeLoache et al., 2010). That the parent instruc- tion condition did not show accelerated learning relative to the video viewing conditions differs from DeLoache et al.’s (2010) findings. This may have been due to the lack of emphasis on verbal labels in this study, or to the more limited dura- tion and frequency of exposure that we employed relative to DeLoache et al. Because these words are frequent in input to children, it may be that the degree of enhanced exposure employed in our study was either insufficient or unnecessary to impact the rate of word learning. However, the dissociation between word learning and gesture learning raises important questions for future research regarding how video learning varies for different learning materials. Of particular interest is whether the visual versus auditory modality impacts the relative efficacy of video versus live instruction (see Brito, Barr, McIntyre, & Simcock, 2012; Simcock et al., 2011, for additional discussion of this issue).

Conclusion

This experimentally controlled, longitudinal investigation reveals that, at least for some stimuli and some video formats, infants exhibit a surprising ability to acquire information obtained via video viewing. These findings bolster Vandewater’s (2011) findings that infants’ communicative repertoire can be expanded through video exposure, and add to the literature by generalizing to sign learning and to production measures. Although the effects were nominally more robust in the parent-supported learning conditions, even infants who viewed vid- eos alone exhibited clear evidence of learning over the course of 3 weeks’ exposure. Because infants viewed multiple presentation formats, we cannot determine how format or formal features (e.g., use of foreground music vs. background music vs. no music) impact learning. Likewise, we cannot deter- mine the optimal duration or frequency of exposure to facilitate learning. This study also does not address the potential risks associated with exposure to media. Nevertheless, we find clear and compel- ling evidence that, at least for baby signs, videos constitute one possible instructional medium for infants.

810 Dayanim and Namy

References

Acredolo, L. P., & Goodwyn, S. (1988). Symbolic gestur- ing in normal infants. Child Development, 59, 450–466. doi:10.2307/1130324

American Academy of Pediatrics. (1999). Media educa- tion. Pediatrics, 104, 341–343. doi:10.1542/peds.2010- 1636

American Academy of Pediatrics. (2011). Media use by children younger than 2 years. Council on Communica- tions and Media. Pediatrics, 128, 1040–1045. doi:10. 1542/peds.2011-1753

Anderson, D. R., Huston, A. C., Schmitt, K. L., Linebar- ger, D. L., & Wright, J. C. (2001). Early childhood tele- vision viewing and adolescent behavior: The recontact study. Monographs of the Society for Research in Child Development, 66(1, Serial No. 264). doi:10.1111/1540- 5834.00120

Anderson, D. R., & Pempek, T. A. (2005). Television and very young children. American Behavioral Scientist, 48, 505–522. doi:10.1177/0002764204271506

Barr, R., & Hayne, H. (1999). Developmental changes in imitation from television during infancy. Child Develop- ment, 70, 1067–1081. doi:10.1111/1467-8624.00079

Barr, R., Muentener, P., & Garcia, A. (2007). Age-related changes in deferred imitation from television by 6- to 18-month-olds. Developmental Science, 10, 910–921. doi:10.1111/j.1467-7687.2007.00641.x

Barr, R., Muentener, P., Garcia, A., Fujimoto, M., & Ch�avez, V. (2007). The effect of repetition on imitation from television during infancy. Developmental Psychobi- ology, 49, 196–207. doi:10.1002/dev.20208

Barr, R., Shuck, L., Salerno, K., Atkinson, E., & Linebar- ger, D. L. (2010). Music interferes with learning from television during infancy. Infant and Child Development, 19, 313–331. doi:10.1002/icd.666

Barr, R., & Wyss, N. (2008). Reenactment of televised con- tent by 2-year olds: Toddlers use language learned from television to solve a difficult imitation problem. Infant Behavior & Development, 31, 696–703. doi:10.1016/ j.infbeh.2008.04.006

Barr, R., Zack, E., Muentener, P., & Garcia, A. (2008). Infants’ attention and responsiveness to television increases with prior exposure and parental interaction. Infancy, 13, 3–56. doi:10.1080/15250000701779378

Brito, N., Barr, R., McIntyre, P., & Simcock, G. (2012). Long-term transfer of learning from books and video during toddlerhood. Journal of Experimental Child Psy- chology, 111, 108–119. doi:10.1016/j.jecp.2011.08.004

Chonchaiya, W., & Pruksananonda, C. (2008). Television viewing associates with delayed language develop- ment. Acta Paediatrica, 97, 977–982. doi:10.1111/j.1651- 2227.2008.00831.x

DeLoache, J. S., Chiong, C., Sherman, K., Islam, N., Van- derborght, M., Troseth, G. L., & O’Doherty, K. (2010). Do babies learn from baby media? Psychological Science, 21, 1570–1574. doi:10.1177/0956797610384145

Fender, J. G., Richert, R. A., Robb, M. B., & Wartella, E. (2010). Parent teaching focus and toddlers’ learning

from an infant DVD. Infant and Child Development, 19, 613–627. doi:10.1002/icd

Fenson, L., Dale, P. S., Reznick, J., Bates, E., Thal, D. J., Pethick, S. J., . . . Stiles, J. (1994). Variability in early communicative development. Monographs of the Society for Research in Child Development, 59(5, Serial No. 242). doi:10.2307/1166093

Fidler, A. E., Zack, E., & Barr, R. (2010). Television view- ing patterns in 6- to 18- month olds: The role of care- giver–infant interactional quality. Infancy, 15, 176–196. doi:10.1111/j.1532-7078.2009.00013.x

Hayne, H., & Herbert, J. (2004). Verbal cues facilitate mem- ory retrieval during infancy. Journal of Experimental Child Psychology, 89, 127–139. doi:10.1016/j.jecp.2004.06.002

Hayne, H., Herbert, J., & Simcock, G. (2003). Imitation from television by 24- and 30-month-olds. Developmen- tal Science, 6, 254–261. doi:10.1111/1467-7687.00281

Khu, M., Graham, S. A., & Ganea, P. A. (2013). Learning from picture books: Infants’ use of naming information. Frontiers In Psychology, 5, Article 144. doi:10.3389/ fpsyg.2014.00144

Krcmar, M. (2011). Word learning in very young children from infant-directed DVDs. Journal of Communication, 61, 780–794. doi:10.1111/j.1460-2466.2011.01561.x

Krcmar, M., Grela, B., & Lin, K. (2007). Can toddlers learn vocabulary from television? An experimental approach. Media Psychology, 10, 41–63. doi:10.1080/ 15213260701300931

Lemish, D. (1987). Viewers in diapers: The early develop- ment of television viewing. In T. R. Lindlof (Ed.), Natu- ral audiences: Qualitative research of media uses and effects (pp. 33–57). Norwood, NJ: Ablex.

Lemish, D., & Rice, M. L. (1986). Television as a talking picture book: A prop for language acquisition. Journal of Child Language, 13, 251–274. doi:10.1017/ S0305000900008047

Linebarger, D. L., & Walker, D. (2005). Infants’ and toddlers’ television viewing and language outcomes. American Behavioral Scientist, 48, 624–645. doi:10.1177/ 0002764204271505

Meltzoff, A. N. (1988a). Imitation of televised models by infants. Child Development, 59, 1221–1229. doi:10.2307/ 1130485

Meltzoff, A. N. (1988b). Infant imitation after a 1-week delay: Long-term memory for novel acts and multiple stimuli. Developmental Psychology, 24, 470–476. doi:10. 1037/0012-1649.24.4.470

Naigles, L. R., & Mayeux, L. (2001). Television as inciden- tal language teacher. In D. G. Singer & J. I. Singer (Eds.), Handbook of children and the media (pp. 135–152). Thousand Oaks, CA: Sage.

Namy, L. L. (2001). What’s in a name when it isn’t a word? 17-month-olds’ mapping of nonverbal symbols to object categories. Infancy, 2, 73–86. doi:10.1207/ S15327078IN0201_5

Namy, L. L., & Waxman, S. R. (1998). Words and sym- bolic gestures: Infants’ interpretations of different forms of symbolic reference. Child Development, 69, 295–308. doi:10.2307/1132165

Parent, M. C. (2013). Handling item-level missing data: Simpler is just as good. The Counseling Psychologist, 41, 568–600. doi:10.1177/0011000012445176

Reiser, R. A., Williamson, N., & Suzuki, K. (1988). Using “Sesame Street” to facilitate children’s recognition of letters and numbers. Educational Communication and Technology Journal, 36, 15–21. Retrieved from http:// www.jstor.org/stable/30218232

Rice, M. L., Huston, A. C., Truglio, R., & Wright, J. (1990). Words from Sesame Street: Learning vocabulary while viewing. Developmental Psychology, 26, 421–428. doi:10.1037/0012-1649.26.3.421

Robb, M., Richert, R., & Wartella, E. (2009). Just a talking book? Word learning from watching baby videos. Brit- ish Journal of Developmental Psychology, 27, 27–45. doi:10. 1348/026151008X320156

Roseberry, S., Hirsh-Pasek, K., Parish-Morris, J., & Go- linkoff, R. M. (2009). Live action: Can young children learn verbs from video? Child Development, 80, 1360– 1375. doi:10.1111/j.1467-8624.2009.01338.x

Rubin, L. H., Witkiewitz, K., St. Andre, J., & Riley, S. (2007). Methods for handling missing data in the behavioral neurosciences: Don’t throw the baby rat out with the bath water. Journal of Undergraduate Neuro- science Education, 5, A71–A77. Retrieved from http:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3592650/

Schmitt, K., & Anderson, D. (2002). Television and reality: Toddler’s use of visual information from video to guide behavior. Media Psychology, 4, 51–76. doi:10.1207/ S1532785XMEP0401_03

Simcock, G., Garrity, K., & Barr, R. (2011). The effect of narrative cues on infants’ imitation from television and picture books. Child Development, 82, 1607–1619. doi:10. 1111/j.1467-8624.2011.01636.x

Singer, J. L., & Singer, D. G. (1998). Barney & Friends as entertainment and education: Evaluating the quality and effectiveness of a television series for preschool children. In J. Asamen & G. L. Berry (Eds.), Research paradigms, television, and social behavior (pp. 305–367). Thousand Oaks, CA: Sage.

Strouse, G. A., O’Doherty, K., & Troseth, G. L. (2013). Effective coviewing: Preschoolers’ learning from video

Infants’ Learning From Videos 811

after a dialogic questioning intervention. Developmental Psychology, 49, 2368–2382. doi:10.1037/a0032463

Strouse, G. A., & Troseth, G. L. (2008). “Don’t try this at home”: Toddlers’ imitation of new skills from people on video. Journal of Experimental Child Psychology, 101, 262–280. doi:10.1016/j.jecp.2008.05.010

Strouse, G. A., & Troseth, G. L. (2014). Supporting toddlers’ transfer of word learning from video. Cognitive Develop- ment, 30, 47–64. doi:10.1016/j.cogdev.2014.01.002

Vaala, S., Linebarger, D. L., Fenstermacher, S., Tedone, A., Brey, E., Barr, R., . . . Calvert, S. L. (2010). Content analysis of language-promoting teaching strategies used in infant directed media? Infant and Child Development, 19, 628–648. doi:10.1002/icd.715

Vandewater, E. A. (2011). Infant word learning from com- mercially available video in the US. Journal of Children and Media, 5, 248–266. doi:10.1080/17482798.2011.584375

Vandewater, E., Barr, R., Park, S. E., & Lee, S. J. (2010). A US study of transfer of learning from video to books in toddlers. Journal of Children and Media, 4, 451–467. doi:10.1080/17482798.2010.510013

Zimmerman, F. J., Christakis, D. A., & Meltzoff, A. N. (2007a). Associations between media viewing and language development in children under age 2 years. Journal of Pediatrics, 151, 364–368. doi:10.1016/j.jpeds. 2007.04.071

Zimmerman, F. J., Christakis, D. A., & Meltzoff, A. N. (2007b). Television and DVD/video viewing in children younger than two years. Archives of Pediatric and Adoles- cent Medicine, 161, 473–479. doi:10.1001/archpedi.161.5. 473

Supporting Information

Additional supporting information may be found in the online version of this article at the publisher’s website:

Appendix S1. Example Pages From Picture Book for the Target Signs “Car,” “Cat,” and “Fish”

Appendix S2. Vocabulary Checklist

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.

Journal ol Personality and Social Psychology 1970, Vol. 16, No. 2, 329-337

ATTENTION IN DELAY OF GRATIFICATION1

WALTER MISCHEL 2 AND EBBE B. EBBESEN

Stanford University

The role of altcntional processes in voluntary delay of reward was explored by manipulating children's attention to the rewards for which they were waiting in a delay-of-gratification paradigm. Preschool children waited for a preferred but delayed reward while facing either the delayed reward, a less preferred but immediately available reward, both rewards, or no rewards. The dependent measure was the amount of time they waited for the preferred outcome before forfeiting it for the sake of the less desired but immediately available one. Results contradicted predictions from psychodynamic theory and from speculations concerning self-instructions during "time binding." Un- expectedly, but in accord with frustrative nonreward theory, voluntary waiting time was substantially increased when subjects could not attend to rewards during the waiting period. Implications are discussed for a theory of the develop- ment of delay of gratification.

The concept of voluntary postponement of immediate gratification for the sake of more distant long-term gains has a central place in conceptualizations of the development of complex human behavior. Formulations stress- ing the role of voluntary delay of reward range from the possible origins of "psychop- athy" and antisocial behavior (e.g., Mowrer & Ullmann, 1945) to characterizations of societal and cultural adaptation patterns in terms of the renunciation of immediate gratifications in favor of disciplined seeking of more substantial future gains. At the em- pirical level, extensive experimental work has been done on delay of reward in animals (e.g., Renner, 1967). Surprisingly, although volun- tary delay behavior has been assumed to be a critical component of such concepts as "ego strength," "impulse control," and "internali- zation," relatively little attention has been devoted to it in empirical work on human social behavior.

One line of research has tried to apply psychoanalytic concepts concerning ego func- tions to motoric inhibition and impulse con- trol (e.g., Singer, 1955). Most of the resulting empirical work has relied on highly indirect

* This study was supported by Research Grant M6830 from the National Institutes of Health, United States Public Health Service. Grateful acknowledg- ment is due to Jerry Zadny for serving as an experi- menter.

2 Requests for reprints should be sent to Walter Mischel, Department of Psychology, Stanford Uni- versity, Stanford, California 94305.

measures of delayed gratification and ego control, mainly inferred from human move- ment responses on the Rorschach (e.g., Spivack, Levine, & Sprigle, 19S9).

In contrast, the present research is part of a larger project to investigate delay of reward with more direct behavioral measures. For example, subjects were required to choose among actual alternatives that varied in delay time and value (e.g., immediate smaller ver- sus delayed but larger rewards) in realistic situations (e.g., Mischel, 1966). Past re- search in this vein has investigated the organi- zation of self-control by exploring the rela- tionship between various preference patterns for immediate smaller rewards or delayed larger rewards and other theoretically rele- vant aspects of personality functioning. The network of associations found here so far indicates, for example, significant relations between preference for delayed rewards and indexes of achievement orientation, social re- sponsibility, age, sociocultural and rearing conditions, and intelligence (e.g., Klineberg, 1968; Mischel, 1961a, 1961b, 1961c; Mischel & Metzner, 1962). Relations have also been found with resistance to temptation (Mischel & Gilligan, 1964) and with severity of psy- chological disturbances (Shybut, 1968). Cor- relational studies were supplemented in re- cent years by experiments to investigate more precisely the determinants of voluntary delay of reward and similar forms of self-control in laboratory situations (e.g., Mischel & Staub,

329

330 WALTEK MISCHEL AND EBBE B. EBBESON

1965; Mischcl, Grusec, & Masters, 1969). As a result of both correlational and experi mental studies, some of the determinants of choice preferences for delayed rewards are be- coming clearer (Mischel, 1966, 1968).

Although choice preferences for immediate or delayed rewards are beginning to be under- stood, the psychological mechanisms through which persons manage to bridge the temporal delay of reward required for attainment of deferred gratification remain remarkably un- studied. In spite of its seemingly evident im- portance, little is known about the self- regulatory mechanisms during the actual de- lay period when the individual must engage in the waiting dictated by his choice of de- layed, larger gratification. Past research has studied verbal choice preferences between re- wards varying in value and in the delay time required to attain them, but just how sub- jects are able to wait during the temporal delay remains unknown. Given that one has chosen to wait for a larger deferred gratifica- tion, how can the delay period be managed? 'The mechanisms that maintain goal-directed delay seem especially important, considering the fact that the ability to sustain self- imposed delay for the sake of larger but delayed consequences appears to be a chief component of most complex higher order human behavior. A main purpose of the pres- ent research, therefore, was to investigate the psychological processes that mediate sustained waiting behavior for delayed gratification.

Freud's (19S9) classic discussion of the transition from primary to secondary process is one of the few theoretical treatments of how delay of gratification may be bridged. According to the psychoanalytic formulation, ideation arises initially when there is a block or delay in the process of direct gratification discharge (Rapaport, 1967, p. 315). During such externally imposed delay, according to Freud, the child constructs a "hallucinatory image" of the physically absent need-satisfy- ing object. Gradually, as a result of repeated association of tension reduction with goal ob- jects, and the development of greater ego organization, the imposed delay of satisfying objects results in the substitution of hallucina- tory satisfactions and other thought processes that convert "free cathexes" into "bound

cathexes" (e.g., Freud, 1959; Singer, 1955). In spite of much psychoanalytic theorizing and speculation about the role of the mental representation of blocked gratifications in the development of delaying capacity, the process remains far from clear.

In their theoretical discussion of impulse control, Jones and Gerard (1967) reasoned that "time-binding," or the capacity to bridge delay of gratification, probably hinges on self-instructional processes through which the individual increases the salience of the de- layed consequences or outcomes of his action. In their view, any factors (situational or within the individual) that make delayed con- sequences more salient should enhance im- pulse control and voluntary delay. Their posi- tion, while emphasizing the self-instructional aspects of attention to deferred outcomes, also implies covert self-reinforcement processes through which the subject may reinforce his own waiting behavior by vividly anticipating some of the positive consequences to which it will lead. Finally, a cognitive-developmental view might lead one to expect that young children may readily forget the delayed out- comes for which they are waiting, and hence cease to wait unless they are reminded of the relevant contingencies and rewards involved in the delay-of-gratification paradigm.

In line with all the foregoing arguments, it seems most plausible that conditions that help the individual to attend mentally to the de- layed reward for which he is waiting should help him to sustain the delay. Operationally, these speculations would suggest that any cues that make the delayed gratification more salient—that help the person to make deferred consequences more psychologically vivid or immediate (e.g., by letting him look at them, by visualizing them in imagination, or by re- minding him of the object for which he is waiting)—should facilitate waiting behavior. Such expectations also seem congruent with the results of earlier work on choice of im- mediate but smaller versus delayed but larger rewards (Maher, 1956; Mischel, 1966; Mischel & Metzner, 1962; Mischel & Staub, 1965). These earlier studies showed that an important determinant of choice preference for delayed rewards is the individual's expec- tation or "trust" that he will really get the

ATTENTION IN D LCI-AY OF GRATIFICATION 331

delayed (hut more valuable) outcome. Con- sequently, conditions that increase the sail ence or visibility of the delayed gratification may enhance the subject's willingness to wait by increasing his subjective probability that the delayed outcome will really materialize and be available after the waiting time ends.

In light of the foregoing considerations, one might expect that voluntary delay behavior is facilitated when the subject converts, as it were, the deferred or delayed object into more tangible form by making it psychologically more immediate, as by providing himself with representations or physical cues about it. The most direct way to increase the salience of the deferred outcomes and to focus attention on them would be to have them physically present and facing the subject, so that he can attend to them readily and vividly. To in- vestigate how attention to delayed and im- mediate outcomes influences waiting behavior for them, a first step would be to manipulate the availability of those outcomes for atten- tion during the delay time.

Previous research on preference for de- layed rewards has been conducted mainly with subjects at least 6 years of age or older. Preliminary observations of the actual waiting behavior of nursery school children suggested, however, that the capacity to wait for long- term goals and to inhibit both immediate gratification and motoric activity seems to de- velop markedly at about ages 3-4. It was hoped, therefore, that research with subjects in this young age range should be especially informative in revealing some of the processes that underlie the genesis of goal-directed waiting.

A first requirement was a paradigm in which such very young children would be willing to remain in an experimental room, waiting entirely alone for at least a short time without becoming upset and debilitat- ingly anxious. As an initial step (after the usual play periods for rapport building) each child was taught a game in which he could immediately summon the experimenter by a simple signal. This step was practiced re- peatedly until the child clearly understood that he could immediately terminate his wait- ing period in the room simply by signaling for the experimenter, who regularly returned

from outside as soon as the child signaled. After this critical procedure had been clearly established, the child was introduced to the relevant contingency. He was shown two ob- jects (e.g., snack-food treats), one of which he clearly preferred (as determined by pre- testing) ; to attain the preferred object he had to wait for it until the experimenter re- turned "by himself." The child was, however, entirely free throughout this waiting period to signal at any time for the experimenter to return; if he signaled, he could have the less preferred object at once, but would forego the more desirable one later.

To manipulate the extent to which children could attend to the reward objects while they were waiting, the rewards were removed from the experimental room in all combinations, creating four conditions with respect to the objects available for attention. In one con- dition, the children waited with both the immediate (less preferred) and the delayed (more preferred) reward facing them in the experimental room, so that they could attend to both outcomes. In another group neither reward was available for the subject's atten- tion, both rewards having been removed from his sight. In the remaining two groups either the delayed reward only or the immediate reward only was left facing the child and available for attention while he waited. The dependent measure was the length of time before each child voluntarily terminated the waiting period.

In accord with the previously discussed theoretical ideas, it was predicted that condi- tions in which the delayed reward was present and visually available would enhance atten- tion to it and hence increase voluntary delay time for it. It was anticipated that the con- dition in which the child was left without either reward would make it most difficult to bridge the delay time and therefore lead to the shortest waiting. In addition it was ex- pected, although less confidently, that the condition in which both the delayed and im- mediate reward were available for attention would best facilitate waiting time. This con- dition might permit the subject to compare and contrast the two outcomes, possibly pro- viding himself with persuasive arguments and self-instructions to help him delay long enough

332 WALTKK MISCHKL AND EBBK B. EBBKSON

Ui achieve his preferred gratification. OH the other hand, one might also plausibly expect maximum delay when the child could focus his attention on the delayed reward without being tempted by the immediate gratifica- tion—that is, the condition in which the de- layed reward was present for attention but the immediate one was not.

METHOD

Subjects and Experimenters The subjects were 16 boys and 16 girls attending

the King Nursery School of Stanford University. Three other subjects were run but eliminated be- cause of their failure to comprehend the instructions as described later. The children ranged in age from 3 years, 6 months, to 5 years, 8 months (with a median age of 4 years, 6 months). The procedures were conducted by two male experimenters. Eight subjects (4 males and 4 females) were assigned ran- domly to each of the four experimental conditions. In each condition each experimenter ran 2 males and 2 females in order to avoid systematic biasing effects from sex or experimenters.

Procedure

The procedures were designed to develop a new method for studying delay behavior experimentally with young subjects. The development of this method was one of the chief goals of the project, and the procedures therefore are described in consider- able detail.

In the week prior to the start of (he experiment, the two male experimenters spent a few days playing with as many children in the nursery school as they could. These nurlurant sessions were designed so that the children would more readily agree to ac- company the experimenters to the "surprise room" and, once there, would be at ease. After obtaining the child's consent to go to the surprise room, the ex- perimenter escorted the child to the experimental room.

The experimental room was a small private cham- ber containing a table, on which lay five -i-inch-long pieces of pretzel and an opaque cake tin. A chair was in front of the table, and on a second chair there was an empty cardboard box. Under the cake tin on the table were five 2-inch-long pretzels and two animal cookies. On the floor near the chair with the cardboard box were four battery-operated toys. On one wall, at right angles with the table, was a one- way mirror. Apart from those objects, the room was empty. The experimenter pointed out the four toys, and before the child could begin to play with the toys, asked the child to sit in the chair which was in front of the table. He then demonstrated each toy briefly in a friendly manner, saying with en- thusiasm after each demonstration that they would play with the toys later on, placing each toy in the cardboard box out of sight of the child. These

references to the toys were designed to help relax the children and also to set up an expectancy that both the child and experimenter would play with the toys sometime later on in the session (thus, ter minating the delay period would not mean having to terminate play in the surprise room).

The next phase required teaching the child the technique for terminating the waiting period and summoning the experimenter at will. For this pur- pose the experimenter said:

Sometimes I have to go out of the room and when 1 do, you can bring me back. Do you sec these tiny pretzels? [The experimenter pointed to the five i-inch pieces of pretzel that would serve as signals.] Well, if I go out of the room and you eat one of these pretzels you can make me come back into the room. You can make me come back! Let's try it. I'll go out of the room now and shut the door. As soon as I do, you eat one of the pretzels and make me come back.

The instructions were repeated, if necessary, until the child seemed to understand them completely.

The experimenter then left the room and shut the door, observing through a small viewing hole in the door when the child ate the pretzel. As soon as the child put the pretzel in his mouth, the experimenter returned, laughing playfully and exclaiming how well the child brought him back into the room. To insure that the child learned reliably how to bring the experimenter back, this sequence was repeated four limes with four of the five small pieces of pretzel, still leaving the last small piece lying next to the as yet unopened cake tin.

Next the experimenter lifted the cake tin, revealing the two sets of reward objects lying there (two cookies and five 2-inch pretzels). The experimenter asked the child which of the two rewards he liked better, and after the child chose, said:

Oh well, you know what? In order for you to eat those [naming the preferred reward] you will have to wait here in your chair and sit very still. I have to go out of the room for a while and when I come back you can eat those [preferred reward] all up. You can take them off the table and eat them right up. But, you know, sometimes, I'm gone a long time and if you want to bring me back you can. Do you know how to bring me back? [All children did know how.| That's right. You eat that little piece [pointing to signal] and I have to come back. But I have to tell you something else. If you eat that and make me come back you can't have [preferred reward]. You can't have them. But you can have all the [naming less preferred reward]! If you sit very still in your chair until 1 come back by myself, then you can eat the [pre- ferred reward]! But if you want to make me come back all you have to do is eat that [pointing to signal] and I'll come back; but then you can't have the [preferred reward]; but you can have all the [less preferred].

ATTKNTION IN DELAY OP GRATIFICATION 33.1

Thus the instructions faced the child with a choice: he could either continue waiting for the more preferred reward until the experimenter re- turned, or he could stop waiting by bringing the experimenter back. If he slopped waiting, then he would receive the less favored (but more im- mediately available) reward and forego the more preferred one. The waiting contingencies were re- peated once more, and then, to assess if the subject understood them, the experimenter asked three ques- tions: "Can you tell me how to bring me back"? "What happens if you eat the pretzel"? "But what happens if you sit very still in your chair and wait for me to come back by myself?" Three chil- dren were unable to answer these questions cor- rectly and were therefore excluded from the data a priori.

At this point the experimenter was informed of the condition in which the subject was to be placed by consulting a slip of paper concealed in the room. This method assured that the experimenter re- mained unaware of the subject's experimental condi- tion until the last possible moment in the procedure. Depending on the condition and the child's choice of preferred reward, the experimenter picked up the cake tin and along with it either nothing, one of the rewards (the more preferred reward or the less preferred reward), or both. The physical arrange- ment was such that the rewards, if left, were directly in front of the child at about shoulder level. In all conditions the signal for summoning the experi- menter was left on the table in front of the child. Thus, depending upon the condition to which the child had been assigned, he was left waiting either with both the delayed and immediate rewards, with either the delayed but more preferred or the im- mediate but less preferred reward, or with neither reward available for attention. Finally, in all con- ditions the experimenter excused himself to leave, and as he was leaving, resummarizcd the waiting in- structions and reminded the child that "no matter what you do, whether you sit and wait for me to come back by myself or whether you bring me back . . . . No matter what you do, we're going to play with my toys when I get back." This instruction was included to stress that the child's waiting behavior would not affect his later play period in the surprise room.

Waiting time was scored from the moment the experimenter shut the door. The experimenter re- turned either as soon as the child signaled or after IS minutes—the criterion time—if the child did not signal. To determine whether or not the child re- membered the waiting contingencies, when the ex- perimenter finally returned he asked the child, "What happens now?" All children answered this question correctly. Subjects were also asked why they had or had not waited. Children who waited to criterion were allowed to cat the chosen, more preferred reward. Those who did not wait to criterion were allowed to eat the unchosen reward. Thereafter each child played with the toys for a while and then was escorted back to his nursery school playroom.

TABLE 1

MEAN MINUTES AND STANDARD DEVIATIONS OF WAITING TIME FOR A DELAYED REWARD

AS A FUNCTION OF ATTENTION

Available for attention

Statistic Both Delayed ImmediateNo rewards rewards reward reward

M 1 1.29 1.03 4.87 5.72 SI) 6.84 2.39 6.57 6.43

RESULTS

In accord with the previously discussed theorizing, it was expected that as the degree of attention paid to the delayed rewards in- creased, the length of time which the children waited would increase. To determine whether or not this prediction was fulfilled, the mean length of time waited (in minutes) was com- puted for each of the four attention condi- tions and is depicted in Table 1. Inspection of these results revealed that unexpectedly, the children waited longest when the rewards were entirely absent—that is, in the condition in which neither the delayed nor the immedi- ate reward was available for attention during the waiting period. Furthermore, the children waited the shortest length of time when both the delayed and the immediate rewards were facing them during the waiting session. These results were exactly opposite to the predic- tions.

An analysis of variance of the mean delay times (Table 2) demonstrated that the over- all effect of attentional conditions was signifi- cant (F = 4.42, df = 3/28, p < .025). To determine the relative contribution of the con- ditions to the overall effect, orthogonal con- trasts were computed (Winer, 1962). The first orthogonal contrast (Ci in Table 2) com- pared the effect of having any reward present for attention with having no reward present during the delay period. This comparison yielded an F of 9.S2 (p < .005, df = 1/28). Thus, children waited much longer for re- wards when the rewards were absent than when any rewards were left available for at- tention. The second orthogonal contrast (r:.) compared mean delay times when both re- wards were present with mean delay times when either the delayed or the immediate re-

334 WALTER MISCHEL AND EBBE B. EBBESON

TABLK 2

ANALYSIS 01? VARIANCE FOR MEAN WAITING TIMES (IN MINUTES) IN EACH

ATTENTION CONDITION

Sourcu

Between C, 6'2 C,

Error

*p < .10. ** p < .025.

*** p < .005.

<V

3 1 1 1

28

MS

144.2 310.5 112.4

9.8 32.63

F

4.4-2** 9.52*** 3.45*

<l

ward was available for attention. The results of this contrast suggested a slight trend to- ward shorter delay when both rewards were present for attention, rather than when only one reward was present (F = 3.45, df — 1/28, p < .1). The final contrast, (Cg), comparing attention to the delayed reward with atten- tion to the immediate reward, was not statis- tically significant (F < 1).

The absolute mean waiting times were prob- ably depressed by the low maximum waiting period used, that is, 15 minutes. Ten subjects out of the total 32 in the study waited the maximum time. Table 3 shows the number of subjects in each condition who waited the full IS minutes. An overall frequency analysis yielded a significant chi-square (x2 = 11.07, p < .025, d/ = 3). Note that not a single child waited the maximum time in the condi- tion in which both rewards were available, whereas 6 out of 8 children waited the maximum time when neither reward was pres- ent. These results further support the findings of the parametric analysis, showing greatest' delay of gratification when the reward ob- jects were not available for attention. In summary, children who were given the oppor- tunity to attend to any of the rewards while

TABLK 3

NUMBER O K CIIII .DKEN WAITING THE M A X I M U M TIMK (15 MINUTES) IN KACII

ATTENTION CONDITION

.Situation

Not wai l ing Waiting

Rewards available foi' aUculion

None Both Delayed

6 2

Immediate

6 2

they were waiting delayed less long than chil- dren who could not attend to any rewards while waiting.

Follow-Up Data

To test the stability of these findings, a partial replication was conducted in later follow-up work. In this replication, the method was altered in one major way. It was recognized that interpretation of the reported results might be somewhat hampered by the fact that the signal for terminating the delay involved eating a tiny pretzel, and that pret- zels also were the rewards. Therefore, instead of the tiny pretzel, a desk bell was used as the signal to terminate the delay period in the follow-up.

Subjects of comparable age from the same nursery school were run in the two conditions that had yielded the main effects. Namely, 12 children were left waiting with neither the delayed nor immediate rewards present and 12 with both rewards present.

The findings clearly supported the previous results. The mean waiting time for the condi- tion in which neither reward was present for attention was 8.9 minutes (SD = 5.26), while the mean waiting time when both rewards were visible was only 3.09 minutes (SD = 5.59). These means were significantly differ- ent in the same direction found previously (t ~ 2.61, dj – 22, p< .025). We therefore may conclude that this attentional condition produced reliable differences in the length of time that children delayed gratification (re- gardless of the signal used to terminate the delay period).

DISCUSSION

Throughout this study unexpected results emerged. A first surprise was the long dura- tion of the waiting periods that many of these young children were able to maintain under some conditions. Tn pilot work, for example, some of the preschool youngsters waited for the preferred reward quietly by themselves, seated alone in a chair for periods sometimes exceeding I hour—an observation that is surprising, considering the widespread belief that young children arc incapable of sustained delay of gratification. Moreover, throughout the entire study not a single child violated

ATTENTION IN DELAY OF GRATIFICATION 335

the stated contingency rule by consuming the preferred but delayed reward before the ex- perimenter's return.

The experimental conditions exerted potent effects on the children's delay behavior, as seen in the finding that six out of eight children waited the maximum 15-minute time when they could attend to neither the imme- diate nor the delayed rewards, whereas the mean waiting time was about 1 minute when they could attend to both rewards. These dif- ferences between conditions suggest that it is inappropriate to conceptualize delay of grati- fication as if it hinged on an all-or-none "ability." Instead, most of the subjects in the present study, in spite of their young age, seemed capable of delay of gratification; the extent to which they did delay depended critically on the specific conditions of the delay period.

The initial theorizing about delay behavior led to predictions of results which were the direct opposite of the obtained findings. It was predicted that attention to the outcomes available in the choice situation while waiting would enhance delay behavior; instead it sharply reduced delay of gratification. Ex- tensive observations of the children's behavior during the delay period provided some clues for a better understanding of the mechanisms through which they mediated their own goal- directed waiting.

One of the most striking delay strategies used by some subjects was exceedingly simple and effective. These children seemed to facili- tate their waiting by converting the aversive waiting situation into a more pleasant non- waiting one. They devised elaborate self- distraction techniques through which they spent their time psychologically doing some- thing (almost anything) other than waiting. Instead of focusing prolonged attention on the objects for which they were waiting, they avoided looking at them. Some children cov- ered their eyes with their hands, rested their heads on their arms, and found other similar techniques for averting their eyes from the reward objects. Many seemed to try to reduce the frustration of delay of reward by gener- ating their own diversions: they talked to themselves, sang, invented games with their hands and feet, and even tried to fall asleep

while waiting—as one child successfully did. These elaborate self-distractions occurred mainly in the rewards-absent condition and almost never in the both-rewards-present con- dition, since in the latter group the children quickly terminated the delay period.

These observations, while obviously incon- clusive, suggest that diverting one's attention away from the delayed reward (while main- taining behavior directed toward its ultimate attainment) may be a key step in bridging temporal delay of reward. That is, learning not to think about what one is awaiting may enhance delay of gratification, much more than does ideating about the outcomes.

These observations also seem consistent with theoretical considerations which (post hoc) could correctly predict the obtained re- sults. Namely, from the perspective of "frtis- trative nonreward" theory (e.g., Amsel, 1958, 1962; Wagner, 1966), the occurrence of non- reward when reward is expected elicits a pri- mary frustration reaction. Congruent with this formulation, when the anticipation of reward is increased, the aversive frustration effect also should be greater. Hence one might pre- dict that cues that enhance the salience of anticipated but still unavailable (delayed) rewards should increase the aversiveness of the delay period. Presumably the greater and more vivid the anticipation of reward, the greater the frustration generated by its delay. This line of reasoning would suggest that con- ditions that decrease the subjects' attention to the blocked reward—and that distract him by internal or overt activity from the frus- trative delay of reward—would make it less aversive for him to continue his goal-directed waiting and hence permit him to wait longer for delayed gratifications. These theoretical expectations seem closely congruent both with the obtained findings and with the more in- formal observations of the children's delay behavior.

The present terminology focuses on the frustrative aspects of not being able to im- mediately obtain the preferred reward in the delay-of-gratification paradigm. The same theoretical considerations, however, apply 1o the aversiveness of the waiting period and of the continuous decisional conflict (between terminating versus waiting longer). In part,

336 WALTER MISCHEL AND EBBE B. EBBESON

attending to the rewards in the waiting par- adigm may be aversive, because it increases the frustration of anticipating the attainment of a blocked reward; in part it may be frus- trative, because it enhances the aversiveness of the waiting situation and accentuates the ongoing decisional conflict. All of these sources of frustration seem an integral part of the delay-of-gratification situation, and at- tention to them makes effective delay be- havior more difficult.

It is of considerable interest that delay be- havior was about the same, regardless of whether the reward in front of the child was the immediately available one or the delayed, more preferred outcome. This finding seems most clearly to contradict any Freudian theo- retical expectations that a mental focus on the delayed outcome (rather than the imme- diate gratification) serves to bridge temporal delay of gratification by providing an internal or "hallucinatory" representation of the de- sired but deferred or blocked outcome.

It might also be thought that the children's waiting behavior in the present situation de- pends on implicit "experimenter demands." Such speculations would predict that the pres- ence of the delayed reward should serve as a cue to the subject that waiting for the de- layed outcome is expected by the experi- menter. Similarly the condition in which only the immediate reward is present should cue less lengthy waiting and enhance willingness to terminate the delay and settle for the im- mediate outcome. These interpretations are untenable, however, because waiting times ;vere similar in the condition in which only the delayed reward was present and the con- dition in which only the immediate reward was present.

One further alternative interpretation that may be suggested is that attention to the rewards simply decreases their subjective value through some sort of habituation pro- cess, and therefore subjects wait less long. In that case one would expect the attention to the delayed reward to result in its subjective devaluation and hence predict shorter waiting when the delayed reward is present, as indeed occurred. The same reasoning, however, also would predict that the presence of the im- mediate reward should lead to its devaluation

and hence generate longer waiting times for the more preferred and absent delayed out- come. The finding that the presence of only the immediate reward in fact led to less delay argues against such a habituation or value- reduction interpretation of the role of atten- tion in delay behavior.

Throughout the present study it has been assumed that the content of subjects' ideation while waiting would be correlated with the attentional conditions to which they were as- signed. Thus it was assumed that making rewards(s) available for attention by facing the subject with them would increase the likelihood that he would actually attend to them during the delay period. While this assumption seems straightforward and parsi- monious, it might conceivably be argued that subjects would actually attend mentally more to the reward objects when the rewards were not physically present than when they were facing them. In that unlikely event, however, one would again have to predict a difference in waiting time between the immediate re- ward only and delayed reward only condi- tions. Presumably subjects would then be fantasizing and thinking more about the ab- sent outcome, which should lead to different waiting times in the immediate reward and delayed reward only attention conditions.

The lack of significant difference in waiting time when the subjects faced the immediate reward or the delayed one does seem under- standable from the perspective of frustrative nonreward theory. When the subject attends to the immediate reward and is tempted to take it, he is frustrated by recalling the con- tingency that attainment of it now prevents his getting the preferred reward later. When the subject attends to the delayed reward, he is frustrated by the fact that he wants it now but cannot have it yet. When he attends to both objects, both of the above aversive frus- trations occur, and hence delay tends to be most difficult—as was the case. In contrast, when the rewards are not visually present for attention, and therefore not. made mentally salient, the subject can more easily avoid the frustration of blocked reward by engaging in various distraction maneuvers both overtly and in his thought processes.

ATTENTION IN DELAY OF GRATIFICATION 337

Thus perhaps the most compelling inter- pretation of the findings may be in terms of the frustrativeness of delay of reward: the presence of the rewards serves to increase the magnitude of the frustration effect and hence decreases delay of gratification by making the waiting period more difficult. The overall findings tentatively suggest that learning to inhibit frustrative ideation, and to divert at- tention away from temptations by focusing, externally and internally, on competing and less frustrating stimuli, may be essential steps for mastery of delay of gratification. If that is true, then the attentional and cognitive processes through which people manage to transform aversive and frustrative conditions into bearable ones by generating their own frustration-reducing distractions become in- triguing questions for future research on self- control. Such research should help us to understand more definitively the mechanisms underlying the present findings.

REFERENCES

AMSEL, A. The role of frustrative nonreward in non- continuous reward situations. Psychological Bul- letin, 19S8, 55, 102-119.

AMSEL, A. Frustrative nonreward in partial rein- forcement and discrimination learning. Psycho- logical Review, 1962, 69, 306-328.

FRETTD, S. (1911) Formulations regarding the two principles in mental functioning. In, Collected papers. Vol. 4. New York: Basic Books, 1959.

JONES, E., & GERARD, H. B. Foundations 0} social psychology. New York: Wiley, 1967.

KLINEBERG, S. L. Future time perspective and the preference for delayed reward. Journal of Per- sonality and Social Psychology, 1968, 8, 253-257.

MAHRER, A. R. The role of expectancy in delayed reinforcement. Journal of Experimental Psychol- ogy, 19S6, 52, 101-105.

MISCHEL, W. Delay of gratification, need for achieve- ment, and acquiescence in another culture. Journal of Abnormal and Social Psychology, 1961, 62, S43-S52. (a)

MISCHEL, W. Father-absence and delay of gratifica- tion: Cross-cultural comparisons. Journal of Ab-

normal and Social Psychology, 1961, 62, 116- 124. (b)

MISCIIEL, W. Preference for delayed reinforcement and social responsibility. Journal of Abnormal and Social Psychology, 1961, 62, 1-7. (c)

MISCHEL, W. Theory and research on the ante- cedents of self-imposed delay of reward. In B. A, Maher (Ed.), Progress in experimental personality research. Vol. 3. New York: Academic Press, 1966.

MISCHEL, W. Personality and assessment. New York: Wiley, 1968.

MiscnuL, W., & GILT.IGAN, C. F. Delay of gratification, motivation for the prohibited gratification, and responses to temptation. Journal of Abnormal and Social Psychology, 1964, 69, 411-417.

MISOIIEL, W., GRUSEC, J., & MASTERS, J. C. Effects of expected delay time on the subjective value of rewards and punishments. Journal of Personality and Social Psychology, 1969,

MISCHEL, W., & METZNER, R. Preference for delayed reward as a function of age, intelligence, and length of delay interval. Journal of Abnormal and Social Psychology, 1962, 64, 425-431.

MISCHKL, W., & STAITU, E. Effects of expectancy on waiting and working for larger rewards. Journal of Personality and Social Psychology, 1965, 2, 625-633.

MOWRER, O. H., & ULLMAN, A. D. Time as a deter- minant in integrative learning. Psychological Re- view, 1945, 52, 61-90.

RAPAPORT, D. On the psychoanalytic theory of thinking. In M. M. Gill (Ed.), The collected papers of David Rapaport. New York: Basic Books, 1967.

RUNNER, K. E. Temporal integration: An incentive approach to conflict resolution. In B. A. Maher (Ed.), Progress in experimental personality re- search. Vol. 4. New York: Academic Press, 1967.

SHYHUT, J. Delay of gratification and severity of psychological disturbance among hospitalized psy- chiatric patients. Journal of Consulting and Clinical Psychology, 1968, 32, 462-468.

SINGER, J. L. Delayed gratification and ego develop- ment: Implications for clinical and experimental research. Journal of Consulting Psychology, 1955, 19, 259-266.

SPIVACK, G., LEVTNE, M., & SPHIGIE, H. Intelligence test performance and the delay function of the ego. Journal of Consulting Psychology, 1959, 23, 428-431.

AGNER, A. R. Frustration and punishment. In R. N. Haber (Ed.), Research on motivation. New York: Holt, Rinehart & Winston, 1966.

(Received June 13, 1969)

W