Biological motion processing in autism spectrum disorders: a behavioural and fMRI investigation

McKay, Lawrie S. (2010) Biological motion processing in autism spectrum disorders: a behavioural and fMRI investigation. PhD thesis, University of Glasgow.

Abstract

There has been much controversy as to whether people with Autism Spectrum Disorders (ASDs) have a specific impairment in processing biological motion, with some studies suggesting there is an impairment (Blake, et. al. 2003; Klin et. al. 2003, Klin & Jones, 2008, Klin et. al. 2009) and others finding that people with ASDs show intact abilities to detect biological motion and categorise actions, but are impaired in emotion categorisation (Moore et. al. 1997; Hubert et. al. 2007, Parron et. al. 2008). Recent studies have found that although behavioural measures of biological motion processing show no differences, adults with ASDs show different patterns of brain activation to controls in response to intact point-light displays (PLDs), with the STS, MT+ and ITG regions showing reduced activity in this population (Herrington et. al. 2007; Parron et. al. 2009). The current thesis aimed to clarify the nature of these difficulties and to try to elucidate the brain regions used to process configural information from PLDs using novel techniques and stimuli.

The first set of experiments were designed to behaviourally test people with ASDs ability to detect biological motion in noise, to categorise actions and to categorise affect from PLDs. Despite finding differences in the two groups in detection of biological motion and affect categorisation in pilot experiments, there were no significant differences between the groups in the main experiments. However, the ASD group showed slightly poorer performance at detecting biological motion and significantly more variability in the action categorisation tasks, suggesting that there may have been an underlying difference between the two groups. Furthermore, an analysis of the pattern of errors tentatively suggested that the ASD group may be using different strategies to categorise affect than controls, particularly for negative affects.

We then devised a novel technique for manipulating the amount of configural information available in a PLD without the need to add different degrees of background noise and used this technique to assess the contribution of configural cues in a direction discrimination task behaviourally and neurally. The results confirmed that in typically developed individuals configural cues significantly improved the participants’ ability to correctly determine the direction of locomotion of a point light walker. Furthermore, the fMRI task found that regions of the inferotemporal, parietal and frontal regions were sensitive to the amount of configural information present in the displays that corresponded to increases in individual participants’ behavioural performance. Lastly, we used the same technique, though with a more powerful fMRI design, to assess the behavioural and neural differences between people with ASDs and controls in response to displays containing different degrees of configural information. We found that both groups were comparable in their ability to discriminate the direction of locomotion from PLDs. However, the brain regions used to process this information were found to be substantially different. In displays in which the configural information enabled participants to accurately judge the direction of locomotion, the control group utilised a similar group of regions as found in the previous experiment. The ASD group showed a pattern of activation suggesting that they predominantly used regions in the temporal and occipital cortex, and more specifically a region in the fusiform gyrus. The results of Granger Causality Mapping analysis, which allows for the mapping of directional to and from seeded regions, confirmed that whereas the control group utilised a network of regions starting from the ITG and connecting to parietal and occipital regions, the ASD group seemed to utilise two separate networks, processing form information in the fusiform gyrus and motion information separately in middle-temporal regions.

The results are discussed in terms of a potential dysfunction of the ITG region in early childhood and two different models of biological motion processing that have been proposed in the recent literature. In TD individuals the model of Giese & Poggio (2003) may be more applicable, in that it proposes the integration of static form cues with motion signals in areas such as the STS. However, a dysfunctional ITG or dysfunctional connections from the ITG to more dorsal regions would disrupt the integration of form and motion processing and force the brain to place additional processing demands on form processing regions in the fusiform gyrus. This would be more in line with the model proposed by Lange and Lappe (2006) in which information can be derived from biological motion in noise without recourse to the actual motion information, through a process of temporal analysis of static postures. Both systems though, may be intact in TD individuals and may share processing requirements depending on the task. Furthermore, it is hypothesised that a dysfunctional ITG may force the brain to place additional demands on regions in the fusiform gyrus and this neural rewiring may be the cause of the developmental delay seen in processing biological motion in people with ASDs (Annaz et. al. 2009). Future studies should examine the roles of the ITG and fusiform area in more detail, both in TD people and in people with ASDs, and determine the specific nature of these neural differences and there behavioural implications for both groups.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Autism spectrum disorders, biological motion, fMRI, Human form, point-light displays.
• Subjects: R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry; B Philosophy. Psychology. Religion > BF Psychology
• Colleges/Schools: College of Science and Engineering > School of Psychology
• Supervisor's Name: Frank, Prof. Pollick and David, Dr. Simmons
• Date of Award: 2010
• Depositing User: Mr Lawrie S McKay
• Unique ID: glathesis:2010-1784
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 12 May 2010
• Last Modified: 20 Oct 2017 12:31
• URI: https://theses.gla.ac.uk/id/eprint/1784