Features and their configuration in face recognition

Tanaka and Farah (1993) have proposed a holistic approach to face recognition in which information about the features of a face and their configuration are combined together in the face representation. An implication of the holistic hypothesis is that alterations in facial configuration should interfere with retrieval of features. In four experiments, the effect of configuration on feature recognition was investigated by creating two configurations of a face, one with eyes close together and one with eyes far apart. After subjects studied faces presented in one of the two configurations (eyes-close or eyes-far), they were tested for their recognition of features shown in isolation, in a new face configuration, and in the old face configuration. It was found that subjects recognized features best when presented in the old configuration, next best in the new configuration, and poorest in isolation. Moreover, subjects were not sensitive to configural information in inverted faces (Experiment 2) or nonface stimuli (i.e., houses; Experiments 3 and 4). Importantly, for normal faces, altering the spatial location of the eyes not only impaired subjects’ recognition of the eye features but also impaired their recognition of the nose and mouth features—features whose spatial locations were not directly altered. These findings emphasize the interdependency of featural and configural information in a holistic face representation.     

任务类型对面孔联合效应和特征效应的调节

采用特征-联合范式的研究发现, 面孔再认阶段可记录到明显的联合效应和特征效应; 然而, 与再认同属情景记忆的另一任务(来源提取)条件的相应效应尚未报道, 背景信息一致性对联合面孔和特征面孔提取的调节作用以及面孔不同特征对特征效应的影响也未考察。为澄清上述问题, 本文采用特征-联合范式, 并以位置为背景展开研究。实验含一个学习任务和两个测验任务(再认和来源提取)。结果显示, 再认和来源提取阶段均记录到显著的联合效应和特征效应, 且两类效应均在来源提取阶段更强; 任务类型与位置背景一致性交互影响提取绩效; 面孔外部特征和内部特征对特征效应的影响相似。表明联合效应和特征效应具有显著的任务类型敏感性, 这些效应是联合面孔和特征面孔的熟悉性较强且对源面孔的回忆加工相对较弱的结果, 且这两类面孔的提取为背景一致性所调节; 任务类型对两类效应的调节与双重加工理论模型相吻合。     

Cognitive mechanisms of false facial recognition in older adults

Older adults show elevated false alarm rates on recognition memory tests involving faces in comparison to younger adults. It has been proposed that this age-related increase in false facial recognition reflects a deficit in recollection and a corresponding increase in the use of familiarity when making memory decisions. To test this hypothesis, we examined the performance of 40 older adults and 40 younger adults on a face recognition memory paradigm involving three different types of lures with varying levels of familiarity. A robust age effect was found, with older adults demonstrating a markedly heightened false alarm rate in comparison to younger adults for "familiarized lures" that were exact repetitions of faces encountered earlier in the experiment, but outside the study list, and therefore required accurate recollection of contextual information to reject. By contrast, there were no age differences in false alarms to "conjunction lures" that recombined parts of study list faces, or to entirely new faces. Overall, the pattern of false recognition errors observed in older adults was consistent with excessive reliance on a familiarity-based response strategy. Specifically, in the absence of recollection older adults appeared to base their memory decisions on item familiarity, as evidenced by a linear increase in false alarm rates with increasing familiarity of the lures. These findings support the notion that automatic memory processes such as familiarity remain invariant with age, while more controlled memory processes such as recollection show age-related decline.     

Associative recognition of face pairs by younger and older adults: the role of familiarity-based processing

Previous research suggested that older adults have a specific impairment in remembering verbal associative information, but it is unclear how elaboration and familiarity might influence this deficit in situations that involve perceptual processing. In the present experiments, younger and older participants studied male-female pairs of faces. Participants were then administered an associative recognition test consisting of previously studied pairs, pairs that contained previously studied items that were not studied together (i.e., conjunction pairs), and entirely new pairs of faces, and participants were instructed to identify pairs that had been presented together at study. Overall, participants were successful at recognizing previously presented pairs but were highly likely to mistakenly endorse conjunction pairs. This pattern was more pronounced for older adults, especially when items were repeated at encoding. Such data suggest that memory for face pairs relies largely on the familiarity of each face and not on a more precise recollection of associative information. (PsycINFO Database Record (c) 2008 APA, all rights reserved).     

Remembering parts and wholes: Configural processing in face recollection

In two experiments, we examined the role of configural processing on states of awareness in face recognition. Configural processing was manipulated by presenting upright and inverted faces during study and test. After studying facial photographs of humans (Experiment 1) and horses (Experiment 2), participants provided a remember/know judgement for each recognised test face. In both experiments, disrupted configural processing had selective effects on states of awareness, so that inversion reduced remember, but not know, responses. Experiment 2 revealed disproportionate inversion effects in that the difference in remember responses between upright and inverted items was significant for human faces but not horses. These findings suggest that configural processing facilitates explicit recollection by providing a distinctive combination of nonsalient event attributes.     

Color and context: An ERP study on intrinsic and extrinsic feature binding in episodic memory

Episodic memory for intrinsic item and extrinsic context information is postulated to rely on two distinct types of representation: object and episodic tokens. These provide the basis for familiarity and recollection, respectively. Electrophysiological indices of these processes (ERP old-new effects) were used together with behavioral data to test these assumptions. We manipulated an intrinsic object feature (color; Experiment 1) and a contextual feature (background; Experiments 1 and 2). In an inclusion task (Experiment 1), the study-test manipulation of color affected object recognition performance and modulated ERP old-new effects associated with both familiarity and recollection. In contrast, a contextual manipulation had no effect, although both intrinsic and extrinsic information was available in a direct feature (source memory) test. When made task relevant (exclusion task; Experiment 2), however, context affected the ERP recollection effect, while still leaving the ERP familiarity effect uninfluenced. We conclude that intrinsic features bound in object tokens are involuntarily processed during object recognition, thus influencing familiarity, whereas context features bound in episodic tokens are voluntarily accessed, exclusively influencing recollection. Figures depicting all the electrodes analyzed are available in an online supplement at www.psychonomic.org/archive.     

Losing face: impaired discrimination of featural and configural information in the mouth region of an inverted face

Given that all faces share the same set of features—two eyes, a nose, and a mouth—that are arranged in similar configuration, recognition of a specific face must depend on our ability to discern subtle differences in its featural and configural properties. An enduring question in the face-processing literature is whether featural or configural information plays a larger role in the recognition process. To address this question, the face dimensions task was designed, in which the featural and configural properties in the upper (eye) and lower (mouth) regions of a face were parametrically and independently manipulated. In a same–different task, two faces were sequentially presented and tested in their upright or in their inverted orientation. Inversion disrupted the perception of featural size (Exp. 1), featural shape (Exp. 2), and configural changes in the mouth region, but it had relatively little effect on the discrimination of featural size and shape and configural differences in the eye region. Inversion had little effect on the perception of information in the top and bottom halves of houses (Exp. 3), suggesting that the lower-half impairment was specific to faces. Spatial cueing to the mouth region eliminated the inversion effect (Exp. 4), suggesting that participants have a bias to attend to the eye region of an inverted face. The collective findings from these experiments suggest that inversion does not differentially impair featural or configural face perceptions, but rather impairs the perception of information in the mouth region of the face.     

Age differences in accuracy and choosing in eyewitness identification and face recognition

Studies of aging and face recognition show age-related increases in false recognitions of new faces. To explore implications of this false alarm effect, we had young and senior adults perform (1) three eye-witness identification tasks, using both target present and target absent lineups, and (2) and old/new recognition task in which a study list of faces was followed by a test including old and new faces, along with conjunctions of old faces. Compared with the young, seniors had lower accuracy and higher choosing rates on the lineups, and they also falsely recognized more new faces on the recognition test. However, after screening for perceptual processing deficits, there was no age difference in false recognition of conjunctions, or in discriminating old faces from conjunctions. We conclude that the false alarm effect generalizes to lineup identification, but does not extend to conjunction faces. The findings are consistent with age-related deficits in recollection of context and relative age invariance in perceptual integrative processes underlying the experience of familiarity.     

Verbal predicates foster conscious recollection but not familiarity of a task-irrelevant perceptual feature – An ERP study

Research on the effects of perceptual manipulations on recognition memory has suggested that (a) recollection is selectively influenced by task-relevant information and (b) familiarity can be considered perceptually specific. The present experiment tested divergent assumptions that (a) perceptual features can influence conscious object recollection via verbal code despite being task-irrelevant and that (b) perceptual features do not influence object familiarity if study is verbal-conceptual. At study, subjects named objects and their presentation colour; this was followed by an old/new object recognition test. Event-related potentials (ERP) showed that a study-test manipulation of colour impacted selectively on the ERP effect associated with recollection, while a size manipulation showed no effect. It is concluded that (a) verbal predicates generated at study are potent episodic memory agents that modulate recollection even if the recovered feature information is task-irrelevant and (b) commonly found perceptual match effects on familiarity critically depend on perceptual processing at study.     

Elaborative processing and conjunction errors in recognition memory

Four experiments were conducted in order to examine the influence of elaborative processing at encoding on recognition memory conjunction lure errors. In these experiments, participants generated cues for compound words as wholes (e.g., haywire) or as separate entities (e.g., hay, wire). Studied words were re-presented in exact form (old) or recombined to form conjunction lures on the recognition test. Participants were asked to make old-new judgments and to indicate whether they had rejected items judged to be new because of recall of a studied item or because of lack of familiarity with an item. The results suggested that recall-to-reject processing and conjunction lure familiarity increased with both types of generation, although generation of cues for compound words as a whole did not influence conjunction lure error rates. An emphasis on processing each constituent of a compound word during encoding increased the familiarity of those constituents more than generation of a compound word as a whole, resulting in an increase in conjunction lure errors. These results suggest that both familiarity and recollection-based monitoring processes influence conjunction lure errors, and therefore support dual-process theories of recognition memory.     

A decrease in conjunction error rates across lags on a continuous recognition task: a robust pattern

In four experiments, the lag retention interval from parent words (e.g., blackmail, jailbird) to a conjunction word (blackbird) was manipulated in a continuous recognition task. Alterations to the basic procedure of Jones and Atchley (2002) were employed in Experiments 1 and 2 to bolster recollection to reject conjunction lures, yet conjunction error rates still decreased across lags of 1 to 20 words. Experiment 3 and a multiexperiment analysis examined the increments of forgetting in familiarity across lags of 1-20 words. Finally, in Experiment 4, participants attempted to identify conjunction probes as "old", and the data were contrasted with those from a previous experiment (Jones & Atchley, 2002, Exp. 1), in which participants attempted not to identify conjunction probes as "old". In support of earlier findings, the decrease in familiarity across lags of 1-20 words appears robust, with a constant level of weak recollection occurring for parent words.     

The face inversion effect—Parts and wholes: Individual features and their configuration

The face inversion effect (FIE) is a reduction in recognition performance for inverted faces (compared to upright faces) that is greater than that typically observed with other stimulus types (e.g., houses). The work of Diamond and Carey, suggests that a special type of configural information, “second-order relational information” is critical in generating this inversion effect. However, Tanaka and Farah concluded that greater reliance on second-order relational information did not directly result in greater sensitivity to inversion, and they suggested that the FIE is not entirely due to a reliance on this type of configural information. A more recent review by McKone and Yovel provides a meta-analysis that makes a similar point. In this paper, we investigated the contributions made by configural and featural information to the FIE. Experiments 1a and1b investigated the link between configural information and the FIE. Remarkably, Experiment 1b showed that disruption of all configural information of the type considered in Diamond and Carey's analysis (both first and second order) was effective in reducing recognition performance, but did not significantly impact on the FIE. Experiments 2 and 3 revealed that face processing is affected by the orientation of individual features and that this plays a major role in producing the FIE. The FIE was only completely eliminated when we disrupted the single feature orientation information in addition to the configural information, by using a new type of transformation similar to Thatcherizing our sets of scrambled faces. We conclude by noting that our results for scrambled faces are consistent with an account that has recognition performance entirely determined by the proportion of upright facial features within a stimulus, and that any ability to make use of the spatial configuration of these features seems to benefit upright and inverted normal faces alike.     

Categorical perception of face identity in noise isolates configural processing.

Neuropsychological evidence suggests that face recognition based on configural (holistic) information can occur in isolation from recognition based on local feature cues. The present study shows that configural processing can be isolated experimentally in normal subjects. A phenomenon is reported that exists only for upright whole faces, namely categorical perception (CP) of face identity in noise. Three discrimination tasks (ABX, better likeness, and similarity ratings) were used to test for perceptual distortion across the category boundary predicted from binary classification of face morphs. Noise was added such that any single local region provided unreliable cues to identity. Under these conditions, CP was found for upright faces but not for inverted faces or single features, even with more than 10,000 trials. The CP-in-noise signature phenomenon was then used to show that configural processing survives image plane rotations of 45 degrees-90 degrees.     

Effects of geometric distortions on face-recognition performance

The importance of 'configural' processing for face recognition is now well established, but it remains unclear precisely what it entails. Through four experiments we attempted to clarify the nature of configural processing by investigating the effects of various affine transformations on the recognition of familiar faces. Experiment 1 showed that recognition was markedly impaired by inversion of faces, somewhat impaired by shearing or horizontally stretching them, but unaffected by vertical stretching of faces to twice their normal height. In experiment 2 we investigated vertical and horizontal stretching in more detail, and found no effects of either transformation. Two further experiments were performed to determine whether participants were recognising stretched faces by using configural information. Experiment 3 showed that nonglobal vertical stretching of faces (stretching either the top or the bottom half while leaving the remainder undistorted) impaired recognition, implying that configural information from the stretched part of the face was influencing the process of recognition--ie that configural processing involves global facial properties. In experiment 4 we examined the effects of Gaussian blurring on recognition of undistorted and vertically stretched faces. Faces remained recognisable even when they were both stretched and blurred, implying that participants were basing their judgments on configural information from these stimuli, rather than resorting to some strategy based on local featural details. The tolerance of spatial distortions in human face recognition suggests that the configural information used as a basis for face recognition is unlikely to involve information about the absolute position of facial features relative to each other, at least not in any simple way.     

Two routes to face perception: evidence from psychophysics and computational modeling

The aim of this study was to separately analyze the role of featural and configural face representations. Stimuli containing only featural information were created by cutting the faces into their parts and scrambling them. Stimuli only containing configural information were created by blurring the faces. Employing an old-new recognition task, the aim of Experiments 1 and 2 was to investigate whether unfamiliar faces (Exp. 1) or familiar faces (Exp. 2) can be recognized if only featural or configural information is provided. Both scrambled and blurred faces could be recognized above chance level. A further aim of Experiments 1 and 2 was to investigate whether our method of creating configural and featural stimuli is valid. Pre-activation of one form of representation did not facilitate recognition of the other, neither for unfamiliar faces (Exp. 1) nor for familiar faces (Exp. 2). This indicates a high internal validity of our method for creating configural and featural face stimuli. Experiment 3 examined whether features placed in their correct categorical relational position but with distorted metrical distances facilitated recognition of unfamiliar faces. These faces were recognized no better than the scrambled faces in Experiment 1, providing further evidence that facial features are stored independently of configural information. From these results we conclude that both featural and configural information are important to recognize a face and argue for a dual-mode hypothesis of face processing. Using the psychophysical results as motivation, we propose a computational framework that implements featural and configural processing routes using an appearance-based representation based on local features and their spatial relations. In three computational experiments (Experiments 4–6) using the same sets of stimuli, we show how this framework is able to model the psychophysical data.     

Memory conjunction errors: The effects of presentation duration and study repetition

The memory-strengthening manipulations of increased presentation duration and increased number of times items were presented were manipulated in the memory conjunction paradigm. Participants viewed parent words once or three times during the study portion of the experiment for 250 ms, 1000 ms, or 3000 ms. After an old/new recognition test participants were asked to give explanations for their answers from the recognition test. The results of true and false recognition as well as recall-to-reject responses (e.g., I know I did not see blackbird since I saw blackmail) indicated that both familiarity and recollection were influenced by the memory-strengthening manipulations. The results provide evidence for dual-process theories of recognition memory and the opposing processes of familiarity and recollection.     

A decrease in conjunction error rates across lags on a continuous recognition task: A robust pattern

In four experiments, the lag retention interval from parent words (e.g., blackmail, jailbird) to a conjunction word (blackbird) was manipulated in a continuous recognition task. Alterations to the basic procedure of Jones and Atchley (2002) were employed in Experiments 1 and 2 to bolster recollection to reject conjunction lures, yet conjunction error rates still decreased across lags of 1 to 20 words. Experiment 3 and a multiexperiment analysis examined the increments of forgetting in familiarity across lags of 1–20 words. Finally, in Experiment 4, participants attempted to identify conjunction probes as “old”, and the data were contrasted with those from a previous experiment (Jones & Atchley, 2002, Exp. 1), in which participants attempted not to identify conjunction probes as “old”. In support of earlier findings, the decrease in familiarity across lags of 1–20 words appears robust, with a constant level of weak recollection occurring for parent words.     

Internal feature saliency as a marker of familiarity and configural processing

Two experiments that explore the internal feature advantage (IFA) in familiar face processing are reported. The IFA involves more efficient processing of internal features for familiar faces over unfamiliar ones. Experiment 1 examined the possibility of a configural basis for this effect through use of a matching task for familiar and unfamiliar faces presented both upright and upside-down. Results revealed the predicted IFA for familiar faces when stimuli were upright, but this was removed when stimuli were inverted. Experiment 2 examined the degree of training required before the IFA was demonstrated. Latency results revealed that whilst 90–180 s of exposure was sufficient to generate an IFA of intermediate magnitude, 180–270 s of exposure was required before the IFA was equivalent to that demonstrated for a familiar face. Taken together, these results offer three conclusions: First, the IFA is reaffirmed as an objective indicator of familiarity; second, the IFA is seen to rest on configural processing; and finally, the development of the IFA with familiarity indicates a development of configural processing with familiarity. As such, insight is gained as to the type of processing changes that occur as familiarity is gradually acquired.     

An own-race advantage for components as well as configurations in face recognition

The own-race advantage in face recognition has been hypothesized as being due to a superiority in the processing of configural information for own-race faces. Here we examined the contributions of both configural and component processing to the own-race advantage. We recruited 48 Caucasian participants in Australia and 48 Chinese participants in Hong Kong, and had them study Caucasian and Chinese faces. After study, they were shown old faces (along with distractors) that were either blurred (isolating configural processing), in which high spatial frequencies were removed from the intact faces, or scrambled (isolating component processing), in which the locations of all face components were rearranged. Participants performed better on the memory test for own-race faces in both the blurred (configural) and scrambled (component) conditions, showing an own-race advantage for both configural and component processing. These results suggest that the own-race advantage in face recognition is due to a general facilitation in different forms of face processing.     

What information underlies correct rejections in short-term recognition memory?

We propose that correct rejections are based on information that contradicts the study set, rather than on insufficient familiarity. Using two-dimensional stimuli, we varied the featural overlap between lures and the study set so that one feature of the lure had occurred during study and the second feature of the lure had not occurred. Familiarity varied with the number of times the studied feature had occurred, whereas detectability of the extralist feature varied with the number of studied alternatives on the same dimension. Lures increased in difficulty with familiarity only when familiarity of the studied feature was confounded with the number of alternatives to the extralist feature. Hence, the difficulty in rejecting the lure was controlled by the number of alternatives to the extralist feature, not by the familiarity of the lure. Current theory requires additional representation assumptions and a new comparison process to accommodate these data.