Georgie Powell1, Catherine R G Jones1, Craig Hedge1, Tony Charman2, Francesca Happé3, Emily Simonoff4, Petroc Sumner1. 1. School of Psychology. 2. Department of Psychology, Institute of Psychiatry Psychology and Neuroscience, King's College London. 3. SGDP Research Centre, Institute of Psychiatry Psychology and Neuroscience, King's College London. 4. Department of Child & Adolescent Psychiatry, Institute of Psychiatry Psychology and Neuroscience, King's College London.
Abstract
OBJECTIVE: Research using cognitive or perceptual tasks in autism spectrum disorder (ASD) often relies on mean reaction time (RT) and accuracy derived from alternative-forced choice paradigms. However, these measures can confound differences in task-related processing efficiency with caution (i.e., preference for speed or accuracy). We examined whether computational models of decision-making allow these components to be isolated. METHOD: Using data from two face-processing tasks (face recognition and egocentric eye-gaze discrimination), we explored whether adolescents with ASD and wide-ranging intellectual ability differed from an age and IQ matched comparison group on model parameters that are thought to represent processing efficiency, caution, and perceptual encoding/motor output speed. RESULTS: We found evidence that autistic adolescents had lower processing efficiency and caution but did not differ from nonautistic adolescents in the time devoted to perceptual encoding/motor output. These results were more consistent across tasks when we only analyzed participants with IQ above 85. Cross-task correlations suggested that processing efficiency and caution parameters were relatively stable across individuals and tasks. Furthermore, logistic classification with model parameters improved discrimination between individuals with and without ASD relative to classification using mean RT and accuracy. Finally, previous research has found that ADHD symptoms are associated with lower processing efficiency, and we observed a similar relationship in our sample, but only for autistic adolescents. CONCLUSIONS: Together, these results suggest that models of decision-making could provide both better discriminability between autistic and nonautistic individuals on cognitive tasks and also a more specific understanding of the underlying mechanisms driving these differences. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
OBJECTIVE: Research using cognitive or perceptual tasks in autism spectrum disorder (ASD) often relies on mean reaction time (RT) and accuracy derived from alternative-forced choice paradigms. However, these measures can confound differences in task-related processing efficiency with caution (i.e., preference for speed or accuracy). We examined whether computational models of decision-making allow these components to be isolated. METHOD: Using data from two face-processing tasks (face recognition and egocentric eye-gaze discrimination), we explored whether adolescents with ASD and wide-ranging intellectual ability differed from an age and IQ matched comparison group on model parameters that are thought to represent processing efficiency, caution, and perceptual encoding/motor output speed. RESULTS: We found evidence that autistic adolescents had lower processing efficiency and caution but did not differ from nonautistic adolescents in the time devoted to perceptual encoding/motor output. These results were more consistent across tasks when we only analyzed participants with IQ above 85. Cross-task correlations suggested that processing efficiency and caution parameters were relatively stable across individuals and tasks. Furthermore, logistic classification with model parameters improved discrimination between individuals with and without ASD relative to classification using mean RT and accuracy. Finally, previous research has found that ADHD symptoms are associated with lower processing efficiency, and we observed a similar relationship in our sample, but only for autistic adolescents. CONCLUSIONS: Together, these results suggest that models of decision-making could provide both better discriminability between autistic and nonautistic individuals on cognitive tasks and also a more specific understanding of the underlying mechanisms driving these differences. (PsycINFO Database Record (c) 2019 APA, all rights reserved).