Elopement in Children With Autism Spectrum Disorder

Children with autism spectrum disorder (ASD) fixate longer on cartoon characters and show greater emotion recognition to cartoons when compared to real faces (1). Observational studies have shown some differences in perception of cartoon faces in children with ASD compared with their typically functioning peers (1–4). Face and emotion processing is a function of social functioning. Impairment in social functioning may predict elopement in children with ASD (5). Elopement is defined in the present article as an escape from the caretaker's confines (e.g., home). Elopement rates are estimated to be nearly four times higher in children with ASD when compared with their siblings who do not have such a diagnosis (5). We investigated different strategies in curbing elopement tendencies in children with ASD.

Method

We describe two children with ASD who were diagnosed by mental health practitioners using DSM-IV criteria. The diagnoses were corroborated in the current visits using the Modified Checklist for Autism in Toddlers-Revised (M-CHAT-R) online scoring system. Their degree of social impairment was estimated using a non-standardized version of the Social Responsiveness Scale (SRS), with a 5-point Likert scale. One question for each of the following criteria was given: awareness, cognition, communication, mannerisms, and motivations. A literature search was conducted to investigate cartoon perception and facial recognition differences in children with ASD. Additionally, we reviewed certain behavioral strategies to decrease elopement in children with ASD.

Case 1

“Joseph” is a 13-year-old middle-class Hispanic boy, previously diagnosed with ASD at the age of 7, who was brought into the emergency department by police when found loitering in the streets. He was verbally provocative and violently swinging his hands at the police. As a result, he had to be restrained prior to arrival. In the emergency department, he became more aggravated upon the sight of his parents and had to be restrained for a second time. The M-CHAT-R questionnaire was administered to support the diagnosis of ASD. Upon evaluation, he believed that his parents were trying to harm him by casting a curse upon him. He repeatedly referred to himself as a hero from certain movies such as Percy Jackson. He compared his parents to villainous characters such as “Medusa the evil witch.” A week prior to the current visit, he was brought to the emergency department for elopement from similar beliefs. He had been harboring feelings of paranoia against his parents for the past 3 months. Upon administering the SRS, he scored 16/25, suggesting a high degree of social impairment for questions of social communication, motivation, and autistic mannerisms.

Case 2

“Chad” is a 15-year-old middle-class Caucasian youth, previously diagnosed with ASD, who was brought to the emergency department by police for wandering suspiciously around a mall parking lot and traversing up and down the elevator persistently. In the emergency department, he was chanting out phrases from his favorite television cartoon shows and was clinically dehydrated from his long march on foot. This was the second time he eloped from home in two months. His mother had placed a tracking device on his wrist after the first elopement. The police had difficulty tracking him, owing to range of signal problems. Upon administration of the SRS, he scored 15/25.

Discussion

A retrospective study conducted by Anderson et al. (5) estimated that about 48% of children with ASD elope from a safe environment. The study also reported that children who were more likely to elope had lower intellectual, communication, and developmental quotients of functioning. Another study reported that more than a third of children with ASD who eloped were never or rarely able to communicate their name, address, or phone number (6). A National Autism Association initiative called “Be REDy” concluded that 36% of children with ASD who eloped nearly drowned and a small number exposed themselves to potentially fatal road traffic accidents (7). Therefore, elopement is a serious behavioral quandary.

Early developmental studies in 8-month and 5-month old infants have shown that typically developing infants tend to look longer at photographed real faces than cartoon faces displaying different emotions, suggesting that they respond more to emotion changes in real faces (2, 8). Rosset et al. showed that children with ASD did not fixate on real face photographs for as long as children without ASD (1, 3). Interestingly, children with ASD fixated on cartoon faces as long as children without ASD, suggesting that the former give preference to cartoon faces over real faces. Another study utilizing brain event-related potentials showed that children with ASD were drawn equally toward photographs of objects and those of familiar faces (9). In contrast, children without ASD focused longer on face photographs than on object photographs. This suggests that real faces do not have the same social significance for children with ASD as they do for typically developing children (9). This “non preference” for real faces in children with ASD may contribute to difficulties in social perception and interaction (1–3, 4, 9).

Garner (10) studied a graduated levels program in a 19-year-old male with ASD. This study employed means such as praising the act of asking permission to go out, providing more access to cartoons as a reward for non-elopement, curtailing elopement with time-outs, and telling the subject to “stop” when he attempted to elope. This was achieved over gradual increments in the period of supervision. Depending upon elopement attempts over time, more reward or less punishment was administered, hence being called a “graduated levels program.” Blocking or physically obstructing a child every time he or she eloped was another intervention reported in one case study (11). Blocking showed a decrease in elopement tendency when compared with rewarding non-elopement with more cartoon time alone. Lang et al. (12) showed that giving verbal and physical attention to the child resulted in reduced elopement when compared with physical redirection (blocking) or access to favorite cartoons alone.

Some of the latest interventions studied are the role of robots in behavior modeling. Robots are capable of humanoid activities without the emotional or contextual component. In a systematic review conducted by Diehl et al. (13), two studies showed promising future applications (14, 15). The first study examined the effects of prosocial behavior when interacting with robots. Pushing buttons on the robot was used as a function of destructive behavior in children with ASD. The more the child pushed buttons on the robot, the less interactive the robot became, with the child displaying negative reinforcement, thus, modulating social behavior using robots. The second study examined robot interactions with children during a game of basketball. The robot varied its skill of playing depending upon the ability of the child to interact and play, thus encouraging children to participate in shared play. Hence, in children with ASD who may give preference for objects over real faces (9), robotic intervention studies may be an innovative approach, which requires thorough evaluation over a larger scale.

The role of virtual reality technologies represents a growing area of research within the field of autism (16). Virtual reality technologies represent a simulation of real-world training environments based on computer graphics. These can be useful in developing interactive programs that involve role playing and cartoons. This may be particularly helpful in children with ASD who give social preference to cartoons.

The limitations in the studies by Garner et al., Call et al., and Lang et al. were the presence of only a single participant, which makes generalization difficult (10–12). Robot-based studies and the virtual reality technology study were limited by small sample sizes and selection bias (14–16). Children with ASD involved in the study had a higher level of functioning.

Conclusions

There is evidence that children with ASD give preference to cartoons and objects over real faces. This observation may support the use of cartoons, robots and virtual reality technology to develop social skills, behaviour modulation, and thus control elopement tendencies. Studies that showed a decrease in elopement incorporated graduated levels programs, blocking, rewarding, robots, and virtual reality technology. Such interventions need further investigation over a larger scale to better establish their role in decreasing elopement.

Key Points/Clinical Pearls