Atypical sensory perception in individuals with Autism Spectrum Disorder (ASD) has been widely discussed in the literature, especially regarding apparent hyporeactivity to thermal stimuli. Clinical observations and parental reports frequently mention an “indifference to pain or temperature,” which has led many researchers to propose that these individuals have elevated perceptual thresholds for thermal stimuli. However, the empirical robustness of this hypothesis is limited. The study conducted by Williams et al. (2019) represents one of the most comprehensive efforts to clarify this issue through objective psychophysical measures, and their findings offer a strong critique of the traditional assumption of thermal hyposensitivity in ASD.
Using a standardized psychophysical threshold protocol, the method of limits, the authors assessed heat and cold detection thresholds in a large sample of 142 participants (83 with ASD and 59 with typical development), spanning a wide age range (7 to 54 years). Contrary to the hyposensitivity hypothesis, the results indicated that mean thermal detection thresholds were statistically equivalent between the groups. This equivalence was confirmed through robust statistical tests that controlled for non-normality of data distribution—an aspect often overlooked in previous studies and which can generate false positives.
However, the investigation revealed a significant difference in another dimension: intra-individual variability across trials, measured by the mean Gini deviation (GMD). Individuals with ASD showed greater variability in responses across trials, suggesting an increase in so-called “perceptual noise.” This variability was strongly correlated with thermal detection thresholds, indicating that some of the observed threshold elevations may be attributed to inconsistent responses rather than reduced sensory sensitivity per se.
Another notable finding is that variables such as intelligence quotient (performance IQ), sex, and age were more consistent predictors of elevated thermal thresholds than ASD diagnosis. In particular, lower performance IQ scores correlated with higher thresholds, and after controlling for this variable, the association between diagnosis and threshold became statistically insignificant. This suggests that previous studies reporting thermal hyposensitivity in ASD may have been influenced by uncontrolled differences in IQ.
Interestingly, although participants with ASD more frequently reported behaviors associated with sensory hyporeactivity (e.g., decreased responsiveness to sensory stimuli) through questionnaires, these self-reports did not significantly correlate with experimentally obtained detection thresholds. This dissociation between subjective and objective measures reinforces the hypothesis that the hyporeactivity observed in clinical settings may originate from more complex neurocognitive processes—such as attentional modulation, emotional reactivity, or cognitive representation of stimuli—and not solely from peripheral alterations in sensory detection.
In methodological terms, the study highlights a crucial point: detection thresholds determined by the threshold method depend on the participant’s response speed, which raises the possibility that the observed differences in thresholds are influenced by variations in reaction time. This issue is especially relevant in ASD, given that variations in response time and its variability have been widely documented in the autistic population, often associated with comorbidity with ADHD symptoms.
Therefore, the evidence gathered by Williams et al. (2019) suggests that the thermal hyporeactivity often attributed to individuals with ASD is not due to a genuine decrease in thermal sensitivity, but can be explained by factors such as perceptual variability, cognitive performance, and differences in motor response processes and attention. This implies a need for a critical review of classical interpretations of the sensory bases of ASD, as well as a call for the use of more precise and reaction-time-independent experimental methods in future investigations of sensory perception in this population.
Reference:
WILLIAMS, ZJ et al. Thermal perceptual thresholds are typical in Autism Spectrum Disorder but strongly related to intra-individual response variability. Scientific Reports, vol. 9, no. 1, p. 12595, 2019. DOI: Thermal Perceptual Thresholds are typical in Autism Spectrum Disorder but Strongly Related to Intra-individual Response Variability – Scientific Reports (https://doi.org/10.1038/s41598-019-49103-2).
