The Impact of Lures on Semantic and Visuospatial Analogical Reasoning in Autistic Children

Background:  In some fluid reasoning studies, children complete analogies like ‘‘Hive is to bee as aquarium is to?’’ by selecting the appropriate answer between response options. These can include a lure: an answer related to one item of the analogy, but which doesn’t correctly complete it. Younger typically developing (TD) children have poorer performance than the older ones in analogies containing lures, reflecting their poorer inhibition skills and ability to manipulate multiple relations (Richland et al., 2006). Only Green and colleagues (2016) studied the effects of lures in analogical reasoning in autism spectrum disorder (ASD). They did so using a social analogical reasoning task and found that young and old autistic and TD children’s performances were similarly impacted by lures. Autistic children were found as good as TD children to solve pictured analogical problems of visuospatial and semantic content (Morsanyi & Holyoak, 2010), but the effects of lures in analogies varying exclusively in content has not been studied in ASD.

Objectives:  To investigate how young and old autistic and TD children are affected by the presence of lures in semantic and visuospatial analogical reasoning problems.

Methods:  37 autistic and 42 TD children matched on age (6-13 years; M=9.54, SD=1.95) and on Raven’s Progressive Matrices (M=57.57 percentile, SD=24.50) completed 80 pictured analogical reasoning problems on a computerized task. Problems varied in content (semantic vs visuospatial) and consisted of 2x2 matrices with last entry to be filled with one of the three responses options presented. 20/40 semantic and 13/40 visuospatial analogies contained a lure in their answer choices. For the semantic problems, lures were semantically related to one item from the matrix, while for the visuospatial problems, lures were visually identical to one item from the matrix. Mixed ANOVAs with Content and Lure presence as within-subject factors and Group and Age (6-9, 10-13 years) as between-subject factors were conducted for accuracy and reaction time (RT).

Results:  For both autistic and TD children, visuospatial content and younger age of the participants decreased accuracy and increased RT. In all children, lure presence decreased accuracy, but did not increase RT. Moreover, for the analogies containing a lure, performance was significantly higher for semantic problems (M=75.19, SD=17.09) than for visuospatial problems (M=65.24, SD=24.4) (p<.05), and older children (M=76.76, SD=16.30) were more accurate than the younger ones (M=61.10, SD=18.52) (p<.05). Interestingly, even though no significant difference in accuracy between autistic and TD children was found for any condition of the task, autistic children (M=5403.32, SD=2198.41) were faster than TD children (M=6775.07, SD=1960.45) to correctly solve visuospatial analogies with and without a lure (p<.05).

Conclusions:  Autistic and TD school-age children are similarly affected by the presence of lure during semantic and visuospatial analogical reasoning, bringing further support to intact fluid reasoning skills and development in ASD. Also, the observed faster reaction time of autistic children relative to TD children for solving visuospatial analogies suggests superior visuospatial reasoning skills in autistic children, independently of the presence of a lure.