Includes bibliographical references (leaves 83-89). / This study extends Gibson and Gouws ' (2003) work on threat and intolerance as well as Kuklinski, Riggle, Ottati, Schwarz and Wyer's (1991) work on the influence of emotion on people's tolerance judgements. Method: Participants were randomly assigned to two experimental groups.

Includes bibliographical references. / Deficits in empathy are considered a defining characteristic of autism spectrum disorders (ASDs). A fundamental difficulty these individuals experience is that of relating to others. Impairments in both top-down and bottom-up processes involved in relating are implicated, as these individuals experience difficulties in relating to others on both a cognitive and a visceral level. Research focussing on their more basic difficulties with social reciprocity (i.e., bottom-up processes of relating) is, however, lacking. A well-established social phenomenon, namely cradling bias (i.e., the preference to cradle an infant to the left of the body midline) is argued to be facilitated by a capacity for empathy. Previous studies provide reason to suspect that this phenomenon taps into the innate ability to relate to another; in other words, it taps into basic bottom-up empathic processes. With the goal of drawing attention to the very basic difficulties in relating and social reciprocity pervading ASDs, I investigated whether the universal leftward cradling bias was absent in ASD children. This research consisted of a pilot and a main study. Both studies were cross-sectional comparisons of two groups: an ASD group and a Control group. For both studies, the method employed was quasi-experimental, as participants were divided into groups based on the pre-existing criterion of diagnosis (i.e., ASD and Control). The ASD groups (for both studies) included children diagnosed with low-functioning autism, high-functioning autism, Asperger's syndrome, and pervasive developmental disorder-not otherwise specified. For the pilot, the Control group consisted of only typically developing children, whereas for the main study, the Control group consisted of typically developing as well as mentally handicapped children. Direct systematic observation was employed to compare the occurrence of cradling bias across groups. In the pilot study, 20 ASD children, aged 6-14, and 20 Control children, aged 5-15, were asked to cradle a doll as if it were an infant she/he were trying to put to sleep or soothe on three separate occasions. These participants were perfectly matched on age and gender. In the main study, 53 ASD children, aged 6-16, and 40 Control children, aged 6-15, were asked to cradle a doll on four separate occasions. These participants were matched as closely as possible on age and gender. Regression analyses on both the pilot and the main study data revealed that the universal leftward bias was absent in ASD children. In contrast, a clear leftward bias was present in Control children, both typically developing and mentally handicapped. These group differences were not accounted for by differences in gender, handedness, intellectual and/or executive functioning. Moreover, differences in the quality of the child-doll (i.e., caregiver-infant) interaction in the cradling bias scenario illustrated the very basic social- emotional difficulties experienced by ASD individuals. The cradling bias scenario is one instance where impairments in primitive bottom-up processes of relating in ASDs can be illustrated. Further investigation of these bottom-up difficulties will allow for a more nuanced understanding of the empathy deficits (i.e., social- emotional deficits) by which ASDs are characterised, which in turn has implications for the management and treatment of these individuals.