The Role of Unimodal Feedback Pathways in Gender Perception During Activation of Voice and Face Areas

Why fractions are difficult? Modeling optimal and sub-optimal integration strategies of numerators and denominators by educated adults

Many children and educated adults experience difficulties in understanding and manipulating fractions. In this study, we argue that a major cause of this challenge is rooted in the need to integrate information from two separate informational sources (i.e., denominator and numerator) according to a normative arithmetic rule (i.e., division). We contend that in some tasks, the correct arithmetic rule is replaced by an inadequate (sub-optimal) operation (e.g., multiplication), which leads to inaccurate representation of fractions. We tested this conjecture by applying two rigorous models of information integration : (a) functional measurement (Experiments 1-3) and (b) conjoint measurement (Experiment 4-5) to data from number-to-line and comparative judgment tasks. These allowed us to compare participants' integration strategies with that of an ideal-observer model. Functional measurement analyses on data from the number-to-line task, revealed that participants could represent the global magnitude of proper and improper fractions quite accurately and combine the fractions' components according to an ideal-observer model. However, conjoint measurement analyses on data from the comparative judgment task, showed that most participants combined these fractions' components according to a sub-optimal (saturated) observer model, that is inconsistent with an ideal-observer (additive) model. These results support the view that educated adults are capable of extracting multiple types of representations of fractions depending on the task at-hand. These representations can be either accurate and conform with normative arithmetic or approximated and inconsistent with normative arithmetic. The latter may lead to the observed difficulties people experience with fractions.

Visual biases in evaluation of speakers' and singers' voice type by cis and trans listeners

Introduction

A singer's or speaker's Fach (voice type) should be appraised based on acoustic cues characterizing their voice. Instead, in practice, it is often influenced by the individual's physical appearance. This is especially distressful for transgender people who may be excluded from formal singing because of perceived mismatch between their voice and appearance. To eventually break down these visual biases, we need a better understanding of the conditions under which they occur. Specifically, we hypothesized that trans listeners (not actors) would be better able to resist such biases, relative to cis listeners, precisely because they would be more aware of appearance-voice dissociations.

Methods

In an online study, 85 cisgender and 81 transgender participants were presented with 18 different actors singing or speaking short sentences. These actors covered six voice categories from high/bright (traditionally feminine) to low/dark (traditionally masculine) voices: namely soprano, mezzo-soprano (referred to henceforth as mezzo), contralto (referred to henceforth as alto), tenor, baritone, and bass. Every participant provided voice type ratings for (1) Audio-only (A) stimuli to get an unbiased estimate of a given actor's voice type, (2) Video-only (V) stimuli to get an estimate of the strength of the bias itself, and (3) combined Audio-Visual (AV) stimuli to see how much visual cues would affect the evaluation of the audio.

Results

Results demonstrated that visual biases are not subtle and hold across the entire scale, shifting voice appraisal by about a third of the distance between adjacent voice types (for example, a third of the bass-to-baritone distance). This shift was 30% smaller for trans than for cis listeners, confirming our main hypothesis. This pattern was largely similar whether actors sung or spoke, though singing overall led to more feminine/high/bright ratings.

Conclusion

This study is one of the first demonstrations that transgender listeners are in fact better judges of a singer's or speaker's voice type because they are better able to separate the actors' voice from their appearance, a finding that opens exciting avenues to fight more generally against implicit (or sometimes explicit) biases in voice appraisal.

Conjoint measurement of perceived transparency and perceived contrast in variegated checkerboards

One fundamental question in vision research is how the retinal input is segmented into perceptually relevant variables. A striking example of this segmentation process is transparency perception, in which luminance information in one location contributes to two perceptual variables: the properties of the transparent medium itself and of what is being seen in the background. Previous work by Robilotto et al. (2002, 2004) suggested that perceived transparency is closely related to perceived contrast, but how these two relate to retinal luminance has not been established. Here we studied the relationship between perceived transparency, perceived contrast, and image luminance using maximum likelihood conjoint measurement (MLCM). Stimuli were rendered images of variegated checkerboards that were composed of multiple reflectances and partially covered by a transparent overlay. We systematically varied the transmittance and reflectance of the transparent medium and measured perceptual scales of perceived transparency. We also measured scales of perceived contrast using cut-outs of the transparency stimuli that did not contain any geometrical cues to transparency. Perceptual scales for perceived transparency and contrast followed a remarkably similar pattern across observers. We tested the empirically observed scales against predictions from various contrast metrics and found that perceived transparency and perceived contrast were equally well predicted by a metric based on the logarithm of Michelson or Whittle contrast. We conclude that judgments of perceived transparency and perceived contrast are likely to be supported by a common mechanism, which can be computationally captured as a logarithmic contrast.

Understanding the Impact of Face Masks on the Processing of Facial Identity, Emotion, Age, and Gender

The COVID-19 pandemic has introduced new challenges for governments and individuals. Unprecedented efforts at reducing virus transmission launched a novel arena for human face recognition in which faces are partially occluded with masks. Previous studies have shown that masks decrease accuracy of face identity and emotion recognition. The current study focuses on the impact of masks on the speed of processing of these and other important social dimensions. Here we provide a systematic assessment of the impact of COVID-19 masks on facial identity, emotion, gender, and age. Four experiments (N = 116) were conducted in which participants categorized faces on a predefined dimension (e.g., emotion). Both speed and accuracy were measured. The results revealed that masks hindered the perception of virtually all tested facial dimensions (i.e., emotion, gender, age, and identity), interfering with normal speed and accuracy of categorization. We also found that the unwarranted effects of masks were not due to holistic processes, because the Face Inversion Effect (FIE) was generally not larger with unmasked compared with masked faces. Moreover, we found that the impact of masks is not automatic and that under some contexts observers can control at least part of their detrimental effects.

How facial aging affects perceived gender: Insights from maximum likelihood conjoint measurement

Conjoint measurement was used to investigate the joint influence of facial gender and facial age on perceived gender (Experiment 1) and perceived age (Experiment 2). A set of 25 faces was created, covarying independently five levels of gender (from feminine to masculine) and five levels of age (from young to old). Two independent groups of observers were presented with all possible pairs of faces from this set and compared which member of the pair appeared as more masculine (Experiment 1) or older (Experiment 2). Three nested models of the contribution of gender and age to judgment (i.e., independent, additive, and saturated) were fit to the data using maximum likelihood. The results showed that both gender and age contributed to the perceived gender and age of the faces according to a saturated observer model. In judgments of gender (Experiment 1), female faces were perceived as more masculine as they became older. In judgments of age (Experiment 2), young faces (age 20 and 30) were perceived as older as they became more masculine. Taken together, the results entail that: (a) observers integrate facial gender and age information when judging either of the dimensions, and that (b) cues for femininity and cues for aging are negatively correlated. This correlation exerts stronger influence on female faces, and can explain the success of cosmetics in concealing signs of aging and exaggerating sexually dimorphic features.