Perceptual interactions of facial dimensions in speeded classification and identification.

The representation underlying the identification and classification of semirealistic line drawings taken from a computer model of the face was investigated by using a speeded classification task and an identification task. These data were analyzed by using a multidimensional extension of signal detection theory, within which varieties of perceptual interactions between dimensions within and across stimuli can be characterized. The dimensions of interest here were eye separation, nose length, and mouth width. The response time and accuracy data from the speeded classification task suggest that processing of a given feature did depend on whether other features were present or absent, but given that other features were present, the results strongly support separability (a macrolevel, across-stimulus form of invariance) for all pairs of facial dimensions used. This separability was confirmed by the subsequent identification task. Owing to its greater resolution, the identification task can reveal interactions that might exist at more microlevels of processing. In fact, the identification data did indicate the presence of perceptual dependence between facial dimensions within a stimulus when the dimensions that were varied were close in spatial proximity (i.e., eye separation and nose length). Within the theoretical framework, perceptual dependence can be interpreted as correlated noise between otherwise separate channels (and hence, is logically distinct from separability). This dependence was greatly reduced for dimensions that were more distant (eyes and mouth). The relation between these results and the configural effects that have been observed with faces as stimuli in other studies is discussed.     

Dimensional separability or flexibility of attention? Age trends in perceiving configural stimuli

The hypothesis that perceptual development proceeds from less to greater dimensional separability was tested by giving a speeded classification task to first and fouth graders. In Experiment 1, children sorted line drawings of a cartoon figure into two piles according to his eyes (open or closed) or his mouth (smiling or frowning). First graders exhibited the pattern characteristic of integral dimensions: An interference effect was found when dimensions were combined orthogonally and a facilitation effect when dimensions were correlated, relative to a condition in which only one dimension varied. Fourth graders exhibited an optional separable pattern: There was facilitation in the correlated condition but no interference in the orthogonal condition. These results support the hypothesis that development proceeds toward greater flexibility of attention rather than simply toward increasing separability. A flexible field hypothesis of visual attention was suggested to explain how older children can exercise such optional separability. This hypothesis correctly predicted the asymmetrical relation between dimensions found in Experiments 2 and 3: For both first and fourth graders, the eyes of a cartoon figure (open or closed) and his location (right or left) were nonseparable when the eyes were relevant and separable (or optional separable) when location was relevant.     

The development of perceived structure and attention: evidence from divided and selective attention tasks

Three experiments provide converging evidence for the view that both perceived structure and attention change during the elementary school years. Kindergarteners, second graders, and adults performed three speeded tasks: divided attention to conjunctions of features, selective attention to orthogonal dimensions and selective attention to correlated dimensions. The tasks were performed with sizes and shapes that were either spatially integrated or spatially separated. In the divided attention task, conjunctions were identified as quickly as single features with integrated stimuli at all ages, but conjunctions were identified more slowly than single features with separated stimuli by all age group. In the orthogonal dimensions task, interference was observed with integrated stimuli across ages, but the interference in adult performance was asymmetric. With separated stimuli, interference was gradually eliminated with increasing age. In correlated dimensions tasks, younger children showed a redundancy gain with integrated stimuli, but no gain was observed in the performances of the older subjects. With separated stimuli there was no redundancy gain at any age. These results were interpreted to mean that integrated stimuli are initially perceived as wholes by all subjects, but that features become more accessible with increasing age. Even so, attention remains constrained by stimulus structure. In contrast, separated stimuli are initially perceived as features at all ages, and the improvement in performance with increasing age is attributable to the increasing command of attentional resources that accompanies development. Our discussion of these findings focuses on three issues: multiple trends in perceptual development, the characteristics of an adequate theory of perceptual representation and processing, and a comparison of the separability hypothesis and other developmental accounts of perceptual development.     

The effect of feature displacement on the perception of well-known faces

The effect of feature displacement within two well-known faces (Terry Wogan and Cyril Smith) was examined. Image processing equipment was used to produce stimuli in which the features of an original facial image were displaced to form a number of modified images. This technique was first reported by Haig, in a recognition study in which the effect of feature displacement within unfamiliar faces was investigated. In the present experiment a perceptual judgement task was carried out in which subjects were presented with a number of modified faces and asked to judge how dissimilar these were with respect to an original image. A multi-dimensional scaling analysis of the comparative judgements of the subjects revealed a two-dimensional solution involving displacement of the eyes and mouth. A clear division between up/down and inward/outward displacement within these features (particularly the eyes) was observed. A similar pattern of results was found for both well-known faces. This result indicates that subjects were responding to changes in the facial configuration produced by the different types of feature displacement (horizontal or vertical), as opposed to movement of the features per se. Finally, the results also indicate that the displacement of inner features (mouth, eyes, nose) was more noticeable than displacement of the outer facial features (eg hairline).     

Perceptual and linguistic interactions in speeded classification: tests of the semantic coding hypothesis

We tested the semantic coding hypothesis, which states that cross-modal interactions observed in speeded classification tasks arise after perceptual information is recoded into an abstract format common to perceptual and linguistic systems. Using a speeded classification task, we first confirmed the presence of congruence interactions between auditory pitch and visual lightness and observed Garner-type interference with nonlinguistic (perceptual) stimuli (low-frequency and high-frequency tones, black and white squares). Subsequently, we found that modifying the visual stimuli by (a) making them lexical (related words) or (b) reducing their compactness or figural 'goodness' altered congruence effects and Garner interference. The results are consistent with the semantic coding hypothesis, but only in part, and suggest the need for additional assumptions regarding the role of perceptual organization in cross-modal dimensional interactions.     

Multidimensional Fechnerian Scaling: Perceptual Separability

A new definition of the perceptual separability of stimulus dimensions is given in terms of discrimination probabilities. Omitting technical details, stimulus dimensions are considered separable if the following two conditions are met: (a) the probability of discriminating two sufficiently close stimuli is computable from the probabilities with which one discriminates the projections of these stimuli on the coordinate axes; (b) the psychometric differential for discriminating two sufficiently close stimuli that differ in one coordinate only does not depend on the value of their matched coordinates (the psychometric differential is the difference between the probability of discriminating a comparison stimulus from a reference stimulus and the probability with which the reference is discriminated from itself). Thus defined perceptual separability is analyzed within the framework of the regular variation version of multidimensional Fechnerian scaling. The result of this analysis is that the Fechnerian metric of a stimulus space with perceptually separable dimensions has the structure of a Minkowski power metric with respect to these dimensions. The exponent of this metric equals the psychometric order of the stimulus space, or 1, whichever is greater.     

眼睛区域构型信息与特征信息的跨维共变增益效应及其加工特异性

面孔知觉可能在区域尺度上发生多维信息整合, 但迄今无特异性实验证据。本研究在两个实验中操纵面孔眼睛区域或嘴巴区域的单维构型或特征信息, 测量人们觉察单维变化或跨维共变的敏感度, 以此检测面孔区域尺度上的多维信息整合有何现象与规律, 进而揭示面孔知觉的多维信息整合机制。实验获得3个发现：(1)正立面孔眼睛区域的信息变化觉察呈现出“跨维共变增益效应”, 跨维信息共变觉察的敏感度显著高于任意一种单维信息变化觉察的敏感度; (2)“跨维共变增益效应”只在正立面孔的眼睛区域出现, 在倒置面孔的眼睛区域、正立面孔的嘴巴区域或倒置面孔的嘴巴区域都没有出现, 因此具有面孔区域特异性和面孔朝向特异性; (3)就单维信息变化觉察而言, 眼睛区域的敏感度不会受到面孔倒置的损伤, 但是嘴巴区域的敏感度会受到面孔倒置的显著损伤。综合可知, 面孔知觉确实会发生区域尺度上的信息整合, 它不是普遍性的信息量效应, 而是特异性的眼睛效应(只发生在正立面孔的眼睛区域)。这是面孔整体加工(face holistic processing)在单维信息分辨和多维信息整合之间建立联系的必经环节。这提示我们对全脸多维信息知觉整合的理解需要从传统的面孔整体加工假设升级到以眼睛为中心的层级化多维信息整合算法(a hierarchical algorithm for multi-dimensional information integration)。     

哭泣表情面孔的注意偏向:眼动的证据

研究采用双选择Oddball范式和线索-靶子范式,并结合眼动技术,以微笑、哭泣和中性表情面孔为刺激材料,综合考察哭泣表情面孔在识别和解离过程中的注意偏向。研究发现:在识别阶段,哭泣表情面孔的识别正确率和反应速度都显著优于微笑表情面孔,进一步的兴趣区注视偏向分析发现,哭泣和微笑表情面孔的注视模式既具有一致的规律,又存在细微的差异;在解离阶段,返回抑制受线索表情类型的影响,在有效线索条件下,哭泣表情线索呈现后个体对目标刺激的平均注视时间和眼跳潜伏期都显著短于其它表情线索。表明哭泣表情面孔在识别和解离过程中具有不同的注意偏向表现,在识别阶段表现为反应输出优势和注视模式上的一致性与差异性;在解离阶段表现为有效线索条件下,对目标刺激定位和视觉加工的促进作用。     

Holistic vs. attribute repetition effects in classifying stimuli

The effect of the just previous stimulus on the current response was determined in three speeded classification experiments. When the stimuli were perceptual patterns, and when they were words, there was marked facilitation if the previous stimulus was identical to the current stimulus. When successive perceptual stimuli were composed of identical elements, but in a changed configuration, there was no measureable sequential effect on “same” response times. These results are for integral stimuli. When stimuli are not integral, as with some studies in the literature, it is suggested that each two-dimensional nonintegral stimulus is treated by the observer as two objects. In those cases, responses can be facilitated by repetition of that object on which responses were based on the previous trial, and this can occur when a value of the other dimensions (other objects) has not repeated. Accordingly, the processing of integral and nonintegral stimuli must be considered separately when attempting to evaluate process models. Further, facilitation due to stimulus repetitions does not account for all of the differences in response times between various classification tasks. These additional differences may be attributed to a multidimensional range effect analogous to the range effects found in psychophysical tasks.     

Recognizing famous people

In daily life, face identification requires that the observer select a single representation from hundreds if not thousands in memory. This breadth of choice is nearly impossible to replicate in the laboratory using newly learned faces, especially in the context of a Bubbles experiment (Gosselin & Schyns, 2001). In this study, we obviated this concern by studying the performance of observers in a face-naming task using 210 faces of celebrities. On each trial, we presented a face randomly sampled with Bubbles. We performed least-square multiple linear regressions on the location of the samples and on accuracy to pinpoint the facial features that were used effectively in this task. Correct face identification relied primarily on the eye areas in spatial frequency bands ranging from 4.37 to 70 cycles per face (cpf) and on the mouth and the nose in a spatial frequency band ranging from 8.75 to 17.5 cpf. A comparison with other studies (Caldara et al., 2005; Schyns, Bonnar, & Gosselin, 2002) in which Bubbles was used with a set of 10 newly learned faces revealed that although the eye areas were useful across studies, the mouth area and higher spatial frequencies gained in importance when few newly learned faces were used.     

Circle size and diameter tilt: A new look at integrality and separability

In six experiments using the speeded classification paradigm, we provide evidence that the ostensibly “separable” dimensions of size and orientation can produce patterns of either separability or asymmetric configurality, depending on the spatial arrangement of the stimuli. In all experiments, subjects classified large or small circles contairting a single line in one of two possible orientations. When the line touched the circle’s perimeter, thereby defining the diameter of the circle (Experiments 1–4), asymmetric configurality obtained: Variations in size interfered with classification by orientation, but variations in orientation did not interfere with classification by size, and redundancy gain was weak or absent. When the lines fell completely within (i.e., did not touch) the circles (Experiments 5 and 6), the results were consistent with separability: There was neither redundancy gain nor interference. Taken together, the results add to the growing body of evidence that classification of specific dimensional pairs as separable or integral may be less feasible than identifying the more general conditions that increase or decrease the psychological salience of dimensional structures and facilitate or interfere with selection of optimal processing strategies.     

Processing two dimensions of nonspeech stimuli: the auditory-phonetic distinction reconsidered.

Nonspeech stimuli were varied along two dimensions--intensity and rise time. In a series of speeded classification tasks, subjected were asked to identify the stimuli in terms of one of these dimensions. Identification time for the dimension of rise time increased when there was irrelevant variation in intensity; however, identification of intensity was unaffected by irrelevant variation in rise time. When the two dimensions varied redundantly, identification time decreased. This pattern of results is virtually identical to that obtained previously for stimuli that vary along a linguistic and a nonlinguistic dimension. The present data, taken together with those from other studies using the same stimuli, suggest that the mechanisms underlying the auditory-phonetic distinction should be reconsidered. The results are also discussed in terms of general models of multidimensional information processing.     

Overt and covert object-based attention

To examine the role of perceptual object representations in the control of eye movements and attention, a pair of experiments adapted the object-cuing paradigm of Egly, Driver, and Rafal (1994) to require eye movements. Displays were pairs of adjacent rectangles, each containing two characters. Observers were asked to make a speeded judgment of a target character’s orientation, and a cue was provided prior to target/distractor onset to indicate the target’s likely location. Gaze-contingent presentation of target and distractors was used to demand overt scanning of displays. Eye movements during task performance evinced two forms of object-based effects. First, saccades following fixation on an invalidly cued item were more likely to be made within the cued rectangle than between rectangles. Second, saccades within the cued rectangle were preceded by shorter dwell times than saccades between rectangles. Extrafoveal processing of stimuli within the cued rectangle, however, was not facilitated, suggesting that covert attention was not allocated more densely within the cued than within the uncued object.     

Identification versus discrimination of distinctive features in speech perception

Dichotically presented CV syllables were presented to 24 normal subjects under two contrasting performance requirements: consonant identification and consonant discrimination. In the identification task, subjects made more errors identifying stimuli distinguished by two as opposed to one distinctive feature. Conversely, in the discrimination task, subjects made more errors for stimuli distinguished by one as opposed to two distinctive features. It was proposed that the results of the identification task reflect the degree of information load in short-term memory. The results of the discrimination task, on the other hand, reflect the degree of perceptual similarity between contrasting stimuli. Both results were accounted for within a distinctive feature framework. Analyses of the individual features which comprised one and two feature distinctions demonstrated the perceptual prominence of the feature [voice] in contrast to the two place features [compact] and [grave].     

Dimensional interactions in vibrotactile information processing

The effect of irrelevant dimensional variation on the processing of vibrotactile stimuli was measured. Six observers performed a speeded classification task with stimuli varying along the dimensions of pitch and loudness. Choice reaction times were obtained for stimuli differing on one dimension alone, on two correlated dimensions, or on two orthogonally varied dimensions. Compared to one-dimension performance, reaction times were faster in the correlated condition and slower in the orthogonal condition. In general, these findings agreed with similar experiments in other modalities, with the exception that the effects in this study tended to be stronger for cases in which loudness, rather than pitch, was the relevant dimension. The results are explained in terms of the integrality of pitch and loudness and of the relative discriminability of dimensions.     

General recognition theory extended to include response times: Predictions for a class of parallel systems

General Recognition Theory (GRT; Ashby & Townsend, 1986) is a multidimensional theory of classification. Originally developed to study various types of perceptual independence, it has also been widely employed in diverse cognitive venues, such as categorization. The initial theory and applications have been static, that is, lacking a time variable and focusing on patterns of responses, such as confusion matrices. Ashby proposed a parallel, dynamic stochastic version of GRT with application to perceptual independence based on discrete linear systems theory with imposed noise (Ashby, 1989). The current study again focuses on cognitive/perceptual independence within an identification classification paradigm. We extend stochastic GRT and its implicated methodology for cognitive/perceptual independence, to an entire class of parallel systems. This goal is met in a distribution-free manner and includes all linear and non-linear systems satisfying very general conditions. A number of theorems are proven concerning stochastic forms of independence. However, the theorems all assume the stochastic version of decisional separability. A vital task remains to investigate the consequences of failures of stochastic decisional separability.     

The facilitated processing of threatening faces: an ERP analysis

Threatening, friendly, and neutral faces were presented to test the hypothesis of the facilitated perceptual processing of threatening faces. Dense sensor event-related brain potentials were measured while subjects viewed facial stimuli. Subjects had no explicit task for emotional categorization of the faces. Assessing early perceptual stimulus processing, threatening faces elicited an early posterior negativity compared with nonthreatening neutral or friendly expressions. Moreover, at later stages of stimulus processing, facial threat also elicited augmented late positive potentials relative to the other facial expressions, indicating the more elaborate perceptual analysis of these stimuli. Taken together, these data demonstrate the facilitated perceptual processing of threatening faces. Results are discussed within the context of an evolved module of fear (A. Ohman & S. Mineka, 2001).     

Primacy of dimensions in vibrotactile perception: an evaluation of early holistic models.

Early holistic models of perception presume that stimuli composed of interacting dimensions can be experienced initially as undifferentiated. This view, formalized through recourse to a Euclidean geometry of perceptual space, predicts that the orientation of axes used to create stimulus sets is unimportant to performance in speeded classification. We tested this idea by using the interacting vibrotactile dimensions of pitch and loudness. Despite perceivers' relatively poor experience with these dimensions, we showed that the orientation corresponding to pitch and loudness was unique in vibrotactile perceptual space; subjects classified stimuli more efficiently at this orientation than at other orientations. Certain holistic models also claim that when stimulus differences are small, perceivers can recognize change without distinguishing the kind of change. We tested this idea by using a signal detection analysis of unspeeded same-different decisions. We found that subjects' ability to notice the kind of change equaled their ability to notice the change alone. In view of these results, which indicate that pitch and loudness are primary in vibrotactile perception, we detail a new conception of dimensional interaction.     

Dimensional interaction between color and pitch

In three experiments I investigated the nature of cross-modal dimensional interaction by testing speeded classification of the synesthetically corresponding dimensions of color (white-black) and pitch (high-low). Experiment 1 showed significant Garner interference when these dimensions were varied orthogonally--redundancy gain for positively correlated dimensions and redundancy loss for negatively correlated dimensions. Attributes from synesthetically congruent stimuli were classified faster than attributes from incongruent stimuli (a congruity effect). Experiment 2 tested a perceptual explanation of this interaction (i.e., that color and pitch are configural dimensions) by using Pomerantz's (1986) diagnostic (comparison of selective and divided attention performance). The configurality hypothesis received little support. Experiment 3 examined the effect of optional processes on color and pitch classification. The results suggest that partly strategic and partly mandatory components may constitute overall performance. Three alternative explanations of the color-pitch interaction--perceptual, semantic, and response based--are evaluated in the context of the present results.