Face recognition,pose and ecological validity

This paper describes five experiments with two main aims; first, to assess the effects of variations in photographic facial pose on recognizability. A second, more fundamental aim was to examine the reliability of laboratory findings relevant to face recognition, in more ecologically valid settings, and to develop a methodology to further this aim. In Experiment 1 best performance was achieved with provision of three facial poses, namely full-face, three-quarter and profile views. With a single pose a three-quarter view was the most helpful. Experiments 2 and 3 attempted to evaluate these findings in a real-life setting using the general public or paid volunteers who attempted to identify a live target in a town centre from a previously presented photograph. The general public produced a very poor response, and the volunteers produced a low detection and high false recognition rate. Experiment 4 increased constraints on the experimental environment and produced a reasonable hit rate, but no effects of pose. Experiment 5 was a laboratory study where the presentation of a live target preceded photographic recognition. Effects of pose reappeared in line with results of Experiment 1. These results underline the danger involved in making practical recommendations arising from purely laboratory based research.     

On the perceptual specificity of memory representations

The present paradigm involved manipulating the congruency of the perceptual processing during the study and test phases of a recognition memory task. During each trial, a gaze-contingent window was used to limit the stimulus display to a region either inside or outside a 10 degrees square centred on the participant's point of gaze, constituting the Central and Peripheral viewing modes respectively. The window position changed in real time in concert with changes in gaze position. Four experiments documented better task performance when viewing modes at encoding and retrieval matched than when they mismatched (i.e., perceptual specificity effects). Viewing mode congruency effects were demonstrated with both verbal and non-verbal stimuli. The present research is motivated and discussed in terms of theoretical views proposed in the 1970s including the levels-of-processing framework and the proceduralist viewpoint. In addition, implications for current processing and multiple systems views of memory are outlined.     

The effects of surface detail on object categorization and naming

Three experiments are reported examining the effects of surface colour and brightness/texture gradients (photographic detail) on object classification and naming. Objects were drawn from classes with either structurally similar or structurally dissimilar exemplars. In Experiment 1a, object naming was facilitated by both congruent surface colour and photographic detail, with the effects of these two variables combining under-additively. In addition incongruent colour disrupted naming accuracy. These effects tended to be larger on objects from structurally similar classes than on objects from structurally dissimilar classes. Experiment 1b examined superordinate classification. There were again advantages due to congruent colour and photographic detail on responses to objects from both structurally similar and structurally dissimilar classes. Incongruent colour disrupted classification accuracy on structurally distinct but not structurally similar items. For structurally similar items, the advantages of congruent surface attributes on classification were smaller than on naming, but this was not the case for structurally dissimilar items. Experiment 2 examined subordinate classification of structurally similar objects. Now effects of congruent and incongruent colour, but not of photographic detail, were found. Experiment 3 showed that congruent and incongruent colour effects occur only when the colours occupy the internal surfaces of objects. The results suggest that surface details can affect object recognition and naming, depending upon: (1) the degree to which objects must be differentiated for a correct response to be made, and (2) the nature of the rate-limiting process determining performance.     

On the perceptual specificity of memory representations

The present paradigm involved manipulating the congruency of the perceptual processing during the study and test phases of a recognition memory task. During each trial, a gaze-contingent window was used to limit the stimulus display to a region either inside or outside a 10° square centred on the participant's point of gaze, constituting the Central and Peripheral viewing modes respectively. The window position changed in real time in concert with changes in gaze position. Four experiments documented better task performance when viewing modes at encoding and retrieval matched than when they mismatched (i.e., perceptual specificity effects). Viewing mode congruency effects were demonstrated with both verbal and non-verbal stimuli. The present research is motivated and discussed in terms of theoretical views proposed in the 1970s including the levels-of-processing framework and the proceduralist viewpoint. In addition, implications for current processing and multiple systems views of memory are outlined.     

Failures of retrieval and comparison constrain change detection in natural scenes

In a change detection paradigm, a target object in a natural scene either rotated in depth, was replaced by another object token, or remained the same. Change detection performance was reliably higher when a target postcue allowed participants to restrict retrieval and comparison processes to the target object (Experiment 1). Change detection performance remained excellent when the target object was not attended at change (Experiment 2) and when a concurrent verbal working memory load minimized the possibility of verbal encoding (Experiment 3). Together, these data demonstrate that visual representations accumulate in memory from attended objects as the eyes and attention are oriented within a scene and that change blindness derives, at least in part, from retrieval and comparison failure.     

Recognizing faces in and out of context

Two experiments are reported which examined the influence of context on face recognition accuracy for novel and familiar faces respectively. Context was manipulated by varying the physical background against which the faces appeared. In Experiment I, 80 student subjects observed 18 faces before attempting to recognize them in a sequence of 36 al ternatives , For half the subj ects, the backgrounds changed from study to test, while for the remainder they stayed the same. In addition, for half the subjects, both the pose and expression of the face also changed, while for the others it remained constant. Changes in pose plus expression and context significantly reduced recognition accuracy for the target faces. Experiment 11 used an identical design, except that the faces of celebrities replaced the novel faces. The influence of context was eliminated but the effects of pose and expression were maintained. However, when only faces which were actually identified by subjects were considered, the effects of pose and expression, too, were eliminated. The significance of these findings for theories of contextual memory are discussed.     

Recognizing faces in and out of context

Two experiments are reported which examined the influence of context on face recognition accuracy for novel and familiar faces respectively. Context was manipulated by varying the physical background against which the faces appeared. In Experiment I, 80 student subjects observed 18 faces before attempting to recognize them in a sequence of 36 alternatives. For half the subjects, the backgrounds changed from study to test, while for the remainder they stayed the same. In addition, for half the subjects, both the pose and expression of the face also changed, while for the others it remained constant. Changes in pose plus expression and context significantly reduced recognition accuracy for the target faces. Experiment II used an identical design, except that the faces of celebrities replaced the novel faces. The influence of context was eliminated but the effects of pose and expression were maintained. However, when only faces which were actually identified by subjects were considered, the effects of pose and expression, too, were eliminated. The significance of these findings for theories of contextual memory are discussed.     

Colour specificity in episodic and in perceptual object recognition with enhanced colour impact

In three experiments we investigated the perceptual specificity of explicit (old-new object recognition) and implicit memory (word-picture matching) for colour. In order to enhance the impact of colour on processing, we increased the number of colours per object and we impaired shape information. We presented multicoloured pictures (Experiment 1), blurred and partially occluded pictures (Experiment 2) and coloured line drawings in visual noise (Experiment 3). Experiments 1 and 2 had an intentional study phase; the study phase of Experiment 3 was an incidental colour or category naming task. Changing colour from study to test always had negative effects on episodic recognition although colour was irrelevant. In contrast, in the matching task old pictures were generally matched faster than new ones independent of their colour congruence. In Experiment 3, an additional small advantage of congruent colours and of semantic processing occurred. We conclude that two different memory representations contribute to these tasks. Changes of an achromatic, more abstract representation, that is used in normal object recognition, and a representation of the specific exemplar that includes colour. This latter token is used in episodic recognition as well as in unusual perceptual tasks (Experiment 3).     

Configural encoding in visual feature binding

The role of configural encoding in feature binding was assessed using the change detection paradigm. Experiment 1 studied the relevance of location in colour–shape binding with study–test intervals ranging from 0 to 4100 ms. Location was rendered irrelevant by randomising it from study to test display or was kept unchanged from study to test. Results revealed differences between these two conditions at 0 and 200 ms, but not at 1500, 2800, and 4100 ms, suggesting that location was important for feature binding in the initial stages, but performance at longer study–test intervals was impervious to change in locations. Experiment 2 studied the effect of increasing the study-display duration by comparing performance with display durations of 200, 900, and 1500 ms, at the study test intervals of 0 and 2000 ms. Although, there was a general increase in the level of performance with increasing display duration, the interaction effect obtained in Experiment 1 was replicated. Experiments 3 and 4 studied the effect of sequential presentation of stimuli, otherwise retaining the design and procedure of Experiment 2. Different levels of performance in Experiments 2, 3, and 4 revealed the role of configural encoding in the binding process.     

Tests of the generality of the principle of encoding specificity

Two empirical tests of the principle of encoding specificity are reported. In Experiment I, the normative strength of the cues presented on the input and on the recall trial was varied factorially. To lessen the emphasis on strictly associative learning, only half the items were cued in each phase of the study-recall cycle. Recall was higher when the cues remained the same than when they changed. However, regardless of the condition of input cuing, strong output cues were substantially more effective than weak ones. In Experiment II, the to-be-remembered words were shown in the presence of weak cues on the input trial. Recognition in the context of strong extralist cues was compared with recall to the original input cues. On the test of cued recognition, the target words were either generated by the subjects as free associates or presented to them as items on a test constructed by the experimenter. Contrary to previous findings, recall was not found to be superior to recognition. The phenomena of cue-dependent forgetting that have been interpreted as evidence for the principle of encoding specificity appear to have limited generality.     

The object-line inferiority effect in pigeons

Eight pigeons (Columba livia) were trained to discriminate between diagonal lines presented alone or embedded in a redundant L-shape right-angle form. The stimuli were white and were presented in an environment that was otherwise totally dark. Numerous experiments done with human observers have shown a strong superiority effect when the diagonal lines are embedded in redundant contexts. However, in Experiment 1 ofthe present study, the pigeons discriminated significantly better between the two diagonal lines when presented alone than when they were embedded in the right-angle context. In order to check on the possibility that these results were restricted to the semi-Ganzfeld condition of Experiment 1, a second experiment was done with black stimuli presented on a white background. Results of Experiment 2 also showed a strong distractor effect. The results of the present experiments confirmed the predictions of the Heinemann and Chase model of pattern recognition by pigeons.     

Spacing effects in cued-memory tasks for unfamiliar faces and nonwords

Memory for repeated items improves as the interval between repetitions in a list increases (the spacing effect). This study investigated the spacing effect in recognition memory and in a frequency judgment task for unfamiliar target faces that were repeated in the same or in a different pose during incidental learning. Changing the pose between prime and probe trials reduced perceptual repetition priming in a structural discrimination task and also reduced the spacing effect in a subsequent unexpected recognition memory task. Three further experiments confirmed that the spacing effect inrecognition memory (Experiments 2 and 4) or frequency judgment (Experiment 3) was reduced when the pose was changed between repeated presentations at study. Similarly, with nonwords as targets (Experiment 5), changing the font between repeated occurrences of targets at study removed the spacing effect in a subsequent unexpected recognition memory test. These results are interpreted to support the view that short-term perceptual repetition priming underlies the spacing effect in explicit cued-memory tasks for unfamiliar nonsense material.     

Effectiveness of color in picture recognition memory

Investigations were conducted into the effectiveness of color in picture recognition memory. In the study phase, half of the pictures were presented in color and the other half in black and white. In the test phase, half were presented in the same color mode as the study pictures and the other half in the other mode. In immediate and 1-week-delayed tests, the recognition performance was highest when color pictures were used in both the study and test phases. The recall for the color mode of the study pictures, however, was not as good, even with the color pictures. This suggests that the effectiveness of color in picture recognition is not necessarily due to the memory for colors in the pictures themselves, but is probably due to the distinctiveness of features highlighted by the colors. We also found that in the recall performance for the color mode it was more difficult to detect the deletion of colors than to detect their addition.     

The interference effect of emotional expressions on facial identity recognition in preschool-aged children

The present study aims to explore the influence of facial emotional expressions on pre-scholars' identity recognition was analyzed using a two-alternative forced-choice matching task. A decrement was observed in children's performance with emotional faces compared with neutral faces, both when a happy emotional expression remained unchanged between the target face and the test faces and when the expression changed from happy to neutral or from neutral to happy between the target and the test faces (Experiment 1). Negative emotional expressions (i.e. fear and anger) also interfered with children's identity recognition (Experiment 2). Obtained evidence suggests that in preschool-age children, facial emotional expressions are processed in interaction with, rather than independently from, the encoding of facial identity information. The results are discussed in relationship with relevant research conducted with adults and children.     

Exploring the memory advantage for moving scenes

Recognition memory is better for moving images than for static images (the dynamic superiority effect), and performance is best when the mode of presentation at test matches that at study (the study–test congruence effect). We investigated the basis for these effects. In Experiment 1, dividing attention during encoding reduced overall performance but had little effect on the dynamic superiority or study–test congruence effects. In addition, these effects were not limited to scenes depicting faces. In Experiment 2, movement improved both old–new recognition and scene orientation judgements. In Experiment 3, movement improved the recognition of studied scenes but also increased the spurious recognition of novel scenes depicting the same people as studied scenes, suggesting that movement increases the identification of individual objects or actors without necessarily improving the retrieval of associated information. We discuss the theoretical implications of these results and highlight directions for future investigation.     

Long-term perceptual specificity effects in recognition memory: the transformed pictures paradigm.

The effects of a study/test mismatch in the viewing mode of natural scenes on recognition memory performance were examined. At both encoding and retrieval, scenes were presented either by being divided into quarters that were displayed in a sequential cumulative fashion or by scrolling the images through the screen, thereby gradually revealing the content of the images. Half of the participants were tested immediately after encoding and the other half after 48 hours. For both the immediate and delayed retrieval conditions, better recognition memory was demonstrated when viewing modes matched across study and test than when they mismatched. Implications for current processing and multiple systems views of memory are discussed.     

Carry-over effects in face recognition: Processing mode or attentional window?

This study investigates whether or not the dominant processing mode (global or local) in a Navon task is carried over to a face recognition task when the dominant processing mode (configural or featural) in an encoding phase is manipulated. In Experiment 1, Navon figures that were equal in size to a facial photograph were prepared. Participants performed better on the face recognition task after reading large letters in the Navon figures than after reading small letters when configural processing was required in the encoding phase. In Experiment 2, Navon figures that were equal in size to the parts of a face were prepared. The performance after reading large letters did not differ from that after reading small letters. This suggests that both the dominant processing mode in Navon tasks and the size of Navon figures should be considered when discussing the effects of Navon tasks in face recognition.     

Visual impressions of interactions between objects when the causal object does not move

Stimuli were presented that consisted of a stationary row of black-bordered white rectangles. As observers watched, each rectangle in turn from left to right changed from white to black. The final rectangle did not change colour but moved off from left to right. The sequential colour change suggested motion from left to right, and observers reliably reported a visual impression that this illusory motion kicked or bumped the last rectangle, thereby making it move. The impression was stronger when the sequential colour change was faster, but was not significantly affected by the number of the rectangles in the row (ranging from 2 to 8). These results support the conclusion that neither contact nor motion of a causal object is necessary for a visual impression of causality to occur.     

The effect of changes in posture and clothing on the development of unfamiliar person recognition

In contrast to face recognition little is known about the development of unfamiliar whole‐person recognition. Various cues in a person's appearance may contribute to recognition performance. An age‐related influence of different features on person recognition performance is possible. Here, we investigated two variable features of person recognition, namely posture and clothing, in 4‐year‐olds, 6‐year‐olds, 8‐year‐olds, 10‐year‐olds and adults. The experimental methodology of a short‐term recognition‐test with an immediate four alternative forced choice (4AFC)‐array was used. Targets were shown for 5 seconds. Experiment 1 clearly indicates that a variation of both posture and clothing decreases performance in all age groups compared to a standard, i.e. a no‐change condition. An interaction age×variation of posture and clothing was not observed. In order to untangle the two aspects of posture and clothing, Experiment 2 varied only posture between encoding and recognition. An age‐effect was shown, but no other significant result occurred. Experiment 3 investigated variation of clothing only, and found main effects of clothing and age. Because there was no interaction between the two factors we infer that a change of clothing impairs all subjects' recognition performance. Children's person recognition performance suffers in a way similar to adults' person recognition performance when clothing is varied. In contrast, changing posture does not necessarily decrease recognition accuracy, neither in children nor in adults. No developmental gains were found for dealing with the transformation of posture and clothing between the encoding and recognition phase. Copyright © 2003 John Wiley & Sons, Ltd.