Distinguishing the cause and consequence of face inversion: The perceptual field hypothesis

I published a critical review of the face inversion effect (Rossion, 2008) that triggered a few reactions and commentaries by colleagues in the field ( [Riesenhuber and Wolff, 2009] and [83]). Here, I summarize my original paper and attempt to identify the source of both the agreements and disagreements with my colleagues, as well as other authors, regarding the nature of the face inversion effect. My view is that the major cause of the detrimental effect of inversion on an observer’s performance at individual face recognition is the disruption of a perceptual process. This perceptual process is makes and observer see the multiple features of a whole individual upright face at once. It also makes the percept of a given facial feature highly dependent on the location and identity of the other features in the whole face. The perceptual process is holistic because it is driven by a holistic face representation, derived from visual experience. Hence, an inverted face cannot be perceived holistically: the perceptual field of the observer is constricted for inverted faces, each facial feature having to be processed sequentially, independently, i.e. over a smaller spatial window than the whole face. Consequently, it is particularly difficult to perceive diagnostic cues that involve several elements over a wide space on an inverted face, such as long-range relative distances between features (e.g., relative distance between eyes and mouth), or diagnostic cues that are located far away from usual gaze fixation (e.g., mouth-nose distance or mouth shape when fixating between the eyes). These difficulties are mere consequences of face inversion - the cause being a loss of holistic perception-, and it does not follow that relative distances between internal features are necessarily particularly important to recognize faces, that they should be labeled “configural”, or should be given a specific status at the representational level. I argue that distinguishing the cause and consequence(s) of face inversion this way can provide a parsimonious and yet complete theoretical account of the face inversion effect.     

Failure to loose fear: The impact of cognitive load and trait anxiety on extinction

Exposure therapy is an effective technique for fear reduction. However, whether effective exposure requires attentional allocation to the feared situation remains a debated clinical issue. In the present study, the impact of attention allocation in extinction was investigated in an experimental conditioning study. Through a between-subjects manipulation of cognitive load, we created a condition in which participants could allocate their attention to the feared stimulus during extinction (low load condition), and a condition in which attentional allocation was impaired (high load condition). The influence of cognitive load on extinction was examined by comparing electrodermal responses and verbal ratings for the conditioned stimuli in the two extinction load conditions. The results show less successful extinction in the high load condition than in the low load condition. However, this effect was found only in low anxious participants, and it was prominent only on the skin conductance responses. The present results suggest that extinction is not automatic but requires cognitive resources.     

Effects of contrast polarity in paracontrast masking

The visibility of a target stimulus can be suppressed (inhibition) or increased (facilitation) during paracontrast masking. Three processes have been proposed to be involved in paracontrast masking: brief inhibition, facilitation, and prolonged inhibition (Breitmeyer et al., 2006). Brief inhibition is observed when the mask precedes the target at short stimulus onset asynchronies (SOAs) ranging from -10 to -30 msec, whereas prolonged inhibition is effective up to very large SOAs of -450 msec. Facilitation, enhancement in target visibility, can be observed at SOA values between -20 and -110 msec. We further investigated these processes by changing target-mask spatial separation and the contrast polarity of the mask. Our results show that (1) facilitation weakens when spatial separation between the target and mask is increased or when they have opposite contrast polarity, and (2) brief inhibition turns into facilitation for the opposite-polarity mask, whereas prolonged inhibition does not change significantly. These results suggest a fast inhibition mechanism realized in the contrast-specific center-surround antagonism of classical receptive fields for brief inhibition and a slower, higher level cortical processing that is indifferent to contrast polarity for prolonged inhibition.     

Searching through synaesthetic colors

Synaesthesia can be characterized by illusory colors being elicited automatically when one reads an alphanumeric symbol. These colors can affect attention; synaesthetes can show advantages in visual search of achromatic symbols that normally cause slow searches. However, some studies have failed to find these advantages, challenging the conclusion that synaesthetic colors influence attention in a manner similar to the influence of perceptual colors. In the present study, we investigated 2 synaesthetes who reported colors localized in space over alphanumeric symbols’ shapes. The Euclidian distance in CIE xyY color space between two synaesthetic colors was computed for each specific visual search, so that the relationship between color distance (CD) and efficiency of search could be explored with simple regression analyses. Target-to-distractors color salience systematically predicted the speed of search, but the CD between a target or distractors and the physically presented achromatic color did not. When the synaesthetic colors of a target and distractors were nearly complementary, searches resembled popout performance with real colors. Control participants who performed searches for the same symbols (which were colored according to the synaesthetic colors) showed search functions very similar to those shown by the synaesthetes for the physically achromatic symbols.     

Unconscious processing of color and form in metacontrast masking

Three experiments employed a metacontrast masking procedure to examine the extent and nature of priming effects from visual stimuli not consciously perceived. The results showed effects of unconscious stimuli on subsequent target responses that (1) were more consistent, reliable, and not subject to strategic control, as compared with consciously perceived stimuli (Experiment 1); (2) produced both facilitation and interference of subsequent processing (Experiment 2); and (3) did not influence indirect response-related levels of processing (Experiment 3). These results demonstrate that color and form attributes of unconscious stimuli are sufficiently registered within the visual system to influence behavior, and that some of these unconscious effects occur at early levels of stimulus encoding, prior to higher level perceptual and response-related processes.     

Developmental Pragmatics in Normal and Abnormal Children

We propose a critical review of current theories of developmental pragmatics. The underlying assumption is that such a theory ought to account for both normal and abnormal development. From a clinical point of view, we are concerned with the effects of brain damage on the emergence of pragmatic competence. In particular, the paper deals with direct speech acts, indirect speech acts, irony, and deceit in children with head injury, closed head injury, hydrocephalus, focal brain damage, and autism. Since no single theory covers systematically the emergence of pragmatic capacity in normal children, it is not surprising that we have not found a systematic account of deficits in the communicative performance of brain injured children. In our view, the challenge for a pragmatic theory is the determination of the normal developmental pattern within which different pragmatic phenomena may find a precise role. Such a framework of normal behavior would then permit the systematic study of abnormal pragmatic development.     

Detecting deviations from metronomic timing in music: Effects of perceptual structure on the mental timekeeper

The detectability of a deviation from metronomic timing—of a small local increment in interonset interval (IOI) duration—in a musical excerpt is subject to positional biases, or “timing expectations,” that are closely related to the expressive timing (sequence of IOI durations) typically produced by musicians in performance (Repp, 1992b, 1998c, 1998d). Experiment 1 replicated this finding with some changes in procedure and showed that the perception-performance correlation is not the result of formal musical training or availability of a musical score. Experiments 2 and 3 used a synchronization task to examine the hypothesis that participants’ perceptual timing expectations are due to systematic modulations in the period of a mental timekeeper that also controls perceptual-motor coordination. Indeed, there was systematic variation in the asynchronies between taps and metronomically timed musical event onsets, and this variation was correlated both with the variations in IOI increment detectability (Experiment 1) and with the typical expressive timing pattern in performance. When the music contained local IOI increments (Experiment 2), they were almost perfectly compensated for on the next tap, regardless of their detectability in Experiment 1, which suggests a perceptual-motor feed-back mechanism that is sensitive to subthreshold timing deviations. Overall, the results suggest that aspects of perceived musical structure influence the predictions of mental timekeeping mechanisms, thereby creating a subliminal warping of experienced time.     

Further evidence for Coren and Hakstian’s “Methodological implications of interaural correlation: Count heads not ears” and an alternative correction formula

Coren and Hakstian (1990) identified a serious methodological problem that arises in auditory research because of interaural correlation. When measures from both ears of the subjects are pooled together in an experimental design that assumes independence of measures, there can be spuriously high apparent statistical significance. The present paper provides further evidence in support of Coren and Hakstian’s argument and also derives a formula that effectively corrects inflated test statistics resulting from interaural correlation. This formula is a special case of a more general one that applies in many other experimental contexts in which nonindependence of measures is a problem. We found that statistical tests based on our formula have somewhat greater power to detect differences than the kind of correction method advocated by Coren and Hakstian.     

How the brain understands intention: different neural circuits identify the componential features of motor and prior intentions

In this paper we present theoretical and experimental evidence for a set of mechanisms by which intention is understood. We propose that three basic aspects are involved in the understanding of intention. The first aspect to consider is intention recognition, i.e., the process by which we recognize other people's intentions, distinguishing among different types. The second aspect concerns the attribution of intention to its author: the existence of shared neural representations provides a parsimonious explanation of how we recognize other people's intentions (what they are doing), but in and of itself, is not sufficient to determine who the agent is. Once the intention has been recognized and attributed to an agent, the reasons for, and the aim of, the intention are to be considered. Hence, the third aspect concerns the aim-intention motivating the execution of a certain action. We discuss the neural basis of these three theoretical aspects suggesting a conceptual synthesis.