Abstract:Most previous studies only consider face adaptation when the adaptors are faces. No face adaptation effects are observed when the adaptors and test faces do not belong to the same category. This condition implies that face adaptation is category-selective, suggesting that the adaptation occurs at the sensory processing level. The present study attempts to verify the existence of cross-category face adaptation and examines the perceptual results, conditions, and the mechanism of cross-category face adaptation. A total of 43 college students participated in the study. Among the participants, 12 participated in Experiment 1, 12 in Experiment 2, and 19 in Experiment 3. Experiment 1 explored whether cross-category face adaptation and the effect of duration (50, 100, 200, 400, 800, 1600, and 3200 ms) of adaptors exist. Participants were presented with photographs of the objects with gender feature as adaptors, then they identified the gender of a series of faces that were chosen from a morphing spectrum between some male and female faces (the tasks in Experiments 2 and 3 are similar to Experiment 1), whereas behavioral and electrophysiological aftereffects were recorded as indicators of cross-category face adaptation. In Experiment 2, we explored whether the cross-category aftereffects in Experiment 1 were caused by adaption to the feature based on gender. Photographs with items associated with gender, the corresponding names of these items, and the words to describe gender (i.e.,“male”and“female”) were used as adapters. In Experiment 3, we explored whether awareness of adaptors affects cross-category face adaptation by varying degrees of attention load on adaptors (high, low, or no loads). In Experiment 1, all conditions, except for 50 ms duration of adaptors demonstrated aftereffects, and the aftereffects were inverted U-shaped that reached their maximum when the duration of adaptors was 400 ms. Results of Experiment 2 showed that all three kinds of adapters produced cross-category face adaptation. Therefore, cross-category adaptation occurs when the adaptors and test faces share the same properties in terms of the nature of task. Results of Experiment 3 showed that the magnitude of adaptation effect was greater for the no attention than low attention load conditions, which consequently was greater than the high attention load condition. The adaptation effects between the baseline and high attention load conditions were not significant. Therefore, awareness of adaptors is necessary for cross-category adaptation. In summary, the current study proved that high-level cross-category face adaptation could occur when properties or features of adaptors can be automatically inferred from the adaptors, and the adaptors and test faces share the same properties (in this study, it was gender) in terms of the nature of task.

Adaptively triggered comparisons enhance perceptual category learning: evidence from face learning

Newly learned categories induce pre-attentive categorical perception of faces

Category learning increases discriminability of relevant object dimensions in visual cortex

Temporal characteristics of overt attentional behavior during category learning

Category learning can alter perception and its neural correlates

Passive exposure to task-relevant stimuli enhances categorization learning

Contrastive learning explains the emergence and function of visual category-selective regions

Category Learning Selectively Enhances Representations of Boundary-Adjacent Exemplars in Early Visual Cortex

Cross-category Face Adaptation of Feature Association

Learning-Induced Changes in Attentional Allocation During Categorization: A Sizable Catalog of Attention Change As Measured by Eye Movements

Category Representation in Primary Visual Cortex after Visual Perceptual Learning

Categorical learning revealed in activity pattern of left fusiform cortex.

The role of perceptual similarities in determining the asymmetric mixed-category advantage.

The timing of categorical face perception

Automatic selection of eye tracking variables in visual categorization in adults and infants

Multisensory Information Facilitates the Categorization of Untrained Stimuli.

Attentional Bias in Human Category Learning: The Case of Deep Learning

Contrast-dependent response modulation in convolutional neural networks captures behavioral and neural signatures of visual adaptation

Perceptual learning induces fast processing and efficient representation of face

Criterion shifting in an unfamiliar face-matching task: Effects of base rates, payoffs, and perceptual discriminability.