Recognition Memory in Adults with Autism Spectrum Disorder – the Pupil Old/New Effect

Background:  

Memory in individuals with Autism Spectrum Disorder (ASD) has characteristic areas of strength and difficulty. ASD individuals show greater difficulty on less supported tests like free recall but are similar to typically developing (TD) individuals on more supported procedures such as recognition. Most of this research is however done with adults with average intelligence and language abilities. Research into less verbal or younger individuals is uncommon, partly because of methodological challenges. However, in typical as well as clinical populations, gaze measures are becoming popular as indices of memory function and are a good option to overcome these challenges. One such measure - pupil size - has shown robust effects and breakthroughs in terms of understanding the underlying neurochemistry of memory.

Objectives:  

The aim was to establish whether pupil dilation indices would be sensitive to recognition memory processes in ASD adults. We used a wide range of materials to make the findings as generalizable as possible. We aimed to discover more about the utility of the current paradigm with less verbal ASD populations as well as its potential usefulness in the search for a biomarker for ASD.

Methods:  

Data were available for 27 ASD and 30 TD adults aged between 22-64 years and IQs between 75-136. Participants were asked to study sets of 10 words, pictures, non-words and abstract shapes. Materials were black and white images matched for luminance presented one by one in the centre of a computer screen. After the study phase, participants’ memory was tested with a ‘Yes-No’ recognition memory procedure. Pupil size data were measured with a Tobii TX300 remote eye-tracking screen. We calculated a pupil size ratio (pupil size for test item divided by baseline) to control for natural pupil size fluctuation and differences in pupil size between participants at baseline.

Results:  

Both groups remembered visual materials (pictures and abstract shapes) better than verbal materials (words and non-words) and showed superior memory for meaningful over meaningless materials. We compared participants on a pupil size ratio for old (studied) and new (unstudied) materials and found larger pupil sizes for old materials (M = 1.04, SD = 0.03) compared to new materials (M = 1.01, SD = 0.03) for the TD group (p < .0001, Cohen’s d = 0.89). This pupil old/ new effect was absent for the ASD group (M_old = 1.03, SD_old = 0.04; M_new = 1.03, SD_new = 0.08; p = .80, Cohen’s d = 0.04).

Conclusions:  

The current data suggest different underlying recognition memory mechanisms in ASD. Large effect sizes, consistency in findings as well as the dependent measure’s independence from either language or conscious awareness suggest that the pupil size old/new effect might be a good measure to investigate memory in ASD further, especially in less verbally able individuals. In addition the current results give some indication for possible differences in underlying neurochemistry in ASD and therefore make changes in pupil size in ASD a candidate biomarker in autism research.