When Pronouns Are Points: Investigating Reference to Self and Other in Signing ASD Children

Background:  

Children with autism spectrum disorder (ASD) have difficulty with personal pronouns, especially first- and second-person forms, either reversing them (e.g., Kanner, 1943; Bartak & Rutter, 1974) or avoiding their use altogether, instead referring to themselves or others by name (Lee, et al., 1994). The personal pronouns 'I/me' and 'you' in American Sign Language (ASL) are points with the index finger to self and other, respectively. Thus, unlike spoken language pronouns, they transparently pick out the person to whom they refer. Could this characteristic of sign pronouns aid their acquisition by deaf signing children with ASD? Or, will deaf signing children with ASD reverse or avoid pronouns like hearing ASD children do?

Objectives:  

This study aimed to test the pronoun production skills of deaf ASD children in comparison to a group of typically-developing (TD) deaf children.

Methods:  

Children's receptive language ability and nonverbal IQ were measured using the ASL Receptive Skills Test (Enns, et al, 2013) and the Test of Nonverbal Intelligence, Fourth Edition (TONI-4). ASD diagnosis was verified using the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2; Lord, et al., 2012).

Naturalistic language sample: The ADOS evaluation was coded for self-other reference and pointing behavior.

Elicitation: The experimenter took a picture of the child using an iPad, showed him/her the picture, and asked in ASL "who is this?" The experimenter then showed the child a picture of the experimenter and asked again in ASL "who is this?" 

Results:  

Sixteen deaf children with ASD (5 females, all of deaf parents) and 18 typically-developing (TD) deaf children (11 females, all of deaf parents) participated. The two groups were matched for chronological (ASD: ages 5;1-14;4; M_age=9;9, SD=2;6; TD: ages 6;6-12;9; M_age=9;2, SD=1;10) and mental age (ASD: M_age=9;3, SD=3;4; TD: M_age=9;9, SD=3;4); there were no significant group differences (p=.42 and .63, ns, for chronological and mental age, respectively). On the first-person pronoun task, 16 TD children, but only 5 ASD children, produced the ASL 1st-person pronoun; two TD children and 11 ASD children responded by signing their name. Fisher’s Exact Test revealed a significant difference between the TD and ASD groups, p<.01. On the second-person pronoun task, all 18 TD children, but only 8 ASD children, produced the ASL 2nd-person pronoun. The ASD children fingerspelled the experimenter's name or responded idiosyncratically, signing "boy", "man", or "doctor". The group difference was again significant, p<.01. An analysis of the naturalistic language sample showed that all ASD children produced index-finger points, suggesting that the ability to point was not specifically impaired. However, production of pronouns was strongly correlated with language level, r(32)=.51, p<.01, but not mental age (r(32)=.11, ns) or chronological age (r(32)=.004, ns). No pronoun reversals were found in either the elicitation tasks or naturalistic language samples.

Conclusions:  

Deaf ASD children were found to exhibit pronoun avoidance, just as hearing ASD children do, but not pronoun reversal. Pronoun production was strongly correlated with language level, but not with mental or chronological age. Personal pronouns appear to pose a particular challenge to ASD learners in both sign and speech.