Preserved imitation in contrast to limited free application of comfortable hand actions in intellectually able young adults with an autism spectrum disorder

Abstract

Imitation problems are commonly reported in children with an autism spectrum disorder. However, it has not yet been determined whether imitation problems persist into young adulthood. In this study, we investigated imitation skills of 20 intellectually able young adults with autism spectrum disorder relative to 19 age-matched neurotypical adults. For this purpose, we used a bar-transport task, which evokes the application of the end-state comfort principle. Specifically, we examined whether young adults with autism spectrum disorder imitated the means-end structure of a demonstrator’s bar-transport action with and without application of the end-state comfort principle (imitation task). In addition, we examined whether participants spontaneously applied the end-state comfort principle during a similar bar-transport task (free execution task). Results revealed that young adults with autism spectrum disorder imitated the means-end structure of observed actions to the same degree as neurotypical adults (p = 0.428). In contrast, they applied the end-state comfort principle less often during free executed actions (p = 0.035). Moreover, during these actions, they were slower to place the bar into the penholder (p = 0.023), which contributed to the reduced efficiency of their performance. Findings suggest that imitation abilities of young adults with autism spectrum disorder are preserved and that observing others’ actions might promote more efficient action planning in this population.

Keywords

action planning autism spectrum disorder emulation end-state comfort principle means-end imitation young adult

Imitation is defined as an individual’s capacity to replicate observed behaviour. It involves replicating the demonstrator’s movements (i.e. copying the action means) and effects of their movements (i.e. copying the action result), without necessarily understanding the intention or goal behind the actions (Tomasello et al., 2005). By copying the action means, children learn how to act conventionally in their culture. In contrast, when emulating actions, the observer copies only the effect of the demonstrator’s action and not the pattern of the movement. Observers are attracted by the outcome of the action and use movements from their own existing repertoire to achieve a similar outcome (Tomasello et al., 1993). Emulation has less learning potential than means-end imitation (Tomasello et al., 2005).

Rogers and Pennington (1991) were the first to suggest that impaired imitation skills could be a core neuropsychological feature of autism spectrum disorder (ASD), present from early in development, and with possible cascading effects on social relations, interactions and learning throughout the life span. Moreover, several studies indicated that imitation problems in ASD are specific to copying the demonstrator’s action means (Hobson and Lee, 1999; Jiménez et al., 2014). Results of a comprehensive review including 53 studies on imitation and ASD revealed that individuals with ASD performed on average 0.81 standard deviations (SDs) below individuals without ASD on imitation tasks measuring means and results. However, in copying tasks only measuring results, individuals with ASD showed no deficits, suggesting that their emulation abilities are intact (Edwards, 2014).

Jiménez et al. (2014) implemented the bar-transport paradigm (Figure 1) developed by Rosenbaum et al. (1990) in a copying task to distinguish between emulation and means-end imitation. In this paradigm, participants have to relocate a bar, horizontally positioned on a bar support, by rotating it vertically to one of two possible end locations (e.g. penholders). Importantly, Rosenbaum et al. (1990) demonstrated with their paradigm that healthy adults commonly choose to grasp the bar in an awkward fashion to ensure a comfortable hand posture at the end of the movement (i.e. as a result of anticipatory motor planning). This sensitivity towards comfortable end-postures is termed the ‘end-state comfort (ESC) principle’, and implies that the motor system anticipates future body states and plans final grasp postures prior to movement execution. Hence, the ESC principle refers to performance of any type of action resulting in a comfortable end position, even though the action requires an initial sacrifice of comfort at its beginning execution state (Rosenbaum et al., 1990). For instance, a waiter picking up an overturned glass will preferentially use an initial, uncomfortable grip with his or her thumb pointing downwards, in order to hold the glass comfortably thumb-up at the time when the rotation is completed (Wunsch et al., 2013). Comfortable hand postures always correspond to thumb-up hand positions, whereas uncomfortable hand postures correspond to thumb-down hand positions (Hughes, 1996). This anticipatory planning skill emerges at around the end of preschool years, and increases further throughout childhood, stabilizing at the end of childhood (Wunsch et al., 2013). Jiménez et al. (2014) investigated the imitation performances of children with ASD and neurotypical children matched on verbal, mental and chronological age. The demonstrator performed actions either according to or discordant with the ESC principle. In addition, the task goal had either low salience (i.e. end position of a mono-colour bar) or high salience (i.e. end position of a bicolour bar). For low salience goals, participants had to pay more attention to the specific movements executed by the demonstrator (i.e. the action means) in order to determine which end of the bar had to be inserted into which container, whereas for the high salience goals, the focus was more on the action result itself. Results of Jiménez et al.’s (2014) study revealed that children with ASD reproduced low salience actions less frequently, indicating that they had more difficulty with means-end imitation. The authors concluded that children with ASD are more goal-oriented (i.e. emulation) than action-oriented (i.e. means-end imitation).

Figure 1.

Schematic illustration of the original versions of the bar-transport task with initial position (1) and four possible end positions (1, 2, 3, 4). Right-handed adults use underhand grasps to bring the white end of the bar to both penholders (2, 4) and use overhand grasps to bring the black end of the bar to both penholders (1, 3) (Rosenbaum et al., 1990). Note that a comfortable end position always corresponds to thumb-up end position.

Although imitation impairments are consistently reported in children with ASD, previous work has not yet determined whether these problems are long-lasting and persistent, and, as a consequence, affect learning throughout the life span (Vanvuchelen et al., 2011). Additionally, previous work has not yet determined whether imitation problems in adults with ASD are specific to means-end imitation. The aim of this study was to investigate imitation abilities of intellectually able young adults with ASD compared to age-matched neurotypical adults with focus on the means-end structure of imitation. In line with Jiménez et al. (2014), we implemented an adapted version of the bar-transport task (Rosenbaum et al., 1990) in which we used a mono-colour bar to measure means-end imitation abilities.

Our research paradigm improved on prior bar-transport paradigms in several ways. First, it allowed us to differentiate means-end imitation (i.e. copying both the demonstrator’s action means and result) from emulation (i.e. copying the demonstrator’s result without means) and inaccurate performances (i.e. any action with an incorrect result). Second, we controlled for applications of the ESC principle when participants freely executed the bar-transport task. Third, we analysed the speed of actions as an additional relevant measure of action efficiency (i.e. the time necessary to process observed information, formulate a motor plan and execute a movement) (Stoit et al., 2013). Fourth, we controlled for possible attention problems by measuring participants’ responses on simple visual stimuli without any interference of the ESC principle (Leighton et al., 2008).

We predicted that intellectually able adults with ASD, relative to age-matched peers, would be impaired in means-end imitation. Second, we predicted that they would apply the ESC principle less often during freely executed actions (i.e. reduced action efficiency). Finally, we predicted that their action performances would be slower in both the imitation and free execution tasks.

Methods

Participants

For this study, 47 participants between 17 and 29 years old were recruited: 21 young adults with ASD and 26 age-matched neurotypical young adults. Participants were recruited via local advertisements at school and within university grounds in Flanders (the Dutch speaking part of Belgium, Europe). Advertisements were also distributed via mailing lists for university students with special needs. Participants were included in the study based on the following selection criteria: (1) participants needed to attend mainstream secondary education (e.g. high school) or higher education (e.g. college or university); (2) participants with ASD had to have been diagnosed in a certified clinical centre according to a multidisciplinary clinical consensus classification for ASD (Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association (APA), 1994; Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000); and (3) neurotypical participants had to screen negative for ASD symptoms on the Social Responsiveness Scale for Adults (SRS-A), a self-report screening tool for autism symptoms (Constantino et al., 2003; Noens et al., 2012).

Eight out of the 47 participants were excluded from analyses: two neurotypical adults scored above threshold on the SRS-A, four participants (one ASD and three neurotypical) were excluded due to an apparatus malfunction (a sensor had failed to record their hand actions during the free execution and imitation task), and two neurotypical participants were excluded because they produced actions in mirrored view during the imitation task, and mirroring requires the time-consuming process of rational compensation for movements to be made.

The remaining sample consisted of 39 participants: 20 young adults with ASD (15 male; 5 female) and 19 neurotypical adults (10 male; 9 female). Participants were aged between 17 and 29 years old (ASD: mean chronological age 20.7 years; SD 2.7 years; neurotypical adults: mean chronological age 22.0 years; SD 2.4 years). Groups did not differ in chronological age (t₃₇ = 1.60; p = 0.119).

In the ASD group, the mean chronological age of clinical confirmation of ASD diagnosis was 11.8 years (SD = 6.3 years). Comparisons of both groups on the SRS-A revealed that participants with ASD scored significantly higher than neurotypical adults on the total SRS score (U = 308.5; Z = 3.34; p = 0.001; r = 0.53) as well as on Social Communication (U = 285.5; Z = 2.69; p ⩽ 0.01; r = 0.43), Social Motivation (U = 273.0; Z = 2.34; p = 0.02; r = 0.37), Social Consciousness (U = 340.5; Z = 4.24; p ⩽ 0.001; r = 0.68) and Rigidity and Repetition subscales (U = 338.5; Z = 4.18; p ⩽ 0.001; r = 0.67). It is important to note that only 11 out of 20 participants with ASD scored positive for ASD diagnosis according to the self-report SRS-A. Both ASD subgroups (ASD SRS positive and ASD SRS negative) did not differ in chronological age (t₁₈ = −1.75; p = 0.097).

In order to rule out possible differences as a result of intellectual functioning, subgroup analyses were repeated in a subsample of 15 college and university students with ASD compared to 19 neurotypical adults (i.e. five high-school students with ASD were excluded). Chronological age of the latter subgroups did also not differ significantly (t₃₂ = 1.02; p = 0.316).

This study was approved by the ethics committees of Hasselt University and the University Hospitals of Louvain (Flanders, Belgium) before data collection. All persons gave informed consent prior to their inclusion in the study.

Tasks

This study consisted of three tasks: one experimental task (i.e. an imitation task) and two control tasks (i.e. an attention task and a free execution task). The attention task was performed first, followed by the free execution task and the imitation task, respectively.

The attention task aimed to investigate possible attention problems during a visual 4-choice reaction task without interference of the ESC principle. Both accuracy and speed (reaction time) were measured. Accuracy was defined as button presses that corresponded with position and colour of the stimuli shown on screen.

Participants were seated in front of a 17-inch liquid crystal display (LCD) monitor with both forearms pronated and their index fingers resting in between two buttons: two buttons per index finger. Two penholders were shown on the left and right of the monitor screen. During each trial, a bicolour bar appeared on screen with either a blue or red end up in one of the penholders. The four possible stimulus conditions were randomized across trials (24 trials in total). Before the start of every trial, participants were instructed to keep their index fingers positioned in between the buttons to be able to react as fast and accurately as possible. Responses were made by pressing the corresponding button with respect to the item’s colour and position. For instance, when the bar appeared on screen at the right side with the red end up, participants had to press the red button at the right side, using their right index finger.

The free execution task aimed to investigate participants’ application of the ESC principle (i.e. action efficiency) and speed during free executions of actions. Speed was defined as time ranging from starting to lift the hands from the start position until the moment the bar was placed in a penholder.

Participants were seated in front of a 17-inch LCD monitor with both forearms pronated and with each hand resting on a plate located at the table in front. They watched a video clip in which two penholders were shown on the left or right side of a monitor screen. During each trial, a bicolour bar appeared in one of the penholders with either a blue or red end up. In total, there were 24 trials and 4 counterbalanced conditions, similar to the attention task. Bar equipment similar to that shown in the video (e.g. a bar, support and two penholders) was positioned in front of the participants on the table. The bar was a bicolour white iron dowel (20 cm × 2 cm) with a blue and red end. The bar rested on two square cradles (9 cm high, 4 cm × 4 cm wide, positioned 11 cm apart from each other). The two square penholders were 5 cm high and 5 cm × 5 cm wide, and held a cylindrical container of 2.2 cm in diameter. The penholders were located 14 cm to the left and right of the cradles. Cradles and penholders were 3D printed using acrylonitrile butadiene styrene.

Before each trial, participants were instructed to keep their hands on a sensor plate on the table in front. An interruption of an infrared beam (Kingbright L-53P3C phototransistor, wavelength: 940 nm, response time: 15 µs) signalled that the hands were on the plate. During each trial, participants were instructed to place the bar as quickly into the position as demonstrated in the video clip (e.g. in the left or right penholder with the red or blue end up). After dropping the bar in a penholder, participants were requested to bring their hands immediately back to the plate. At the end of each trial, the experimenter placed the bar back on the support cradles, in a predefined random order with respect to its colour. The random order was kept the same for all participants. In addition, in between trials, there was a break of 20 s to give participants time to put their hands back in the starting position for the upcoming trial. Importantly, participants were unaware of the task intentions regarding action efficiency, since no additional instruction or information on application of the ESC principle was given.

The imitation task aimed to investigate the participants’ use of means-end imitations (i.e. the bar in the correct penholder using the demonstrator’s action means), emulation (i.e. the bar in the correct penholder without using the demonstrator’s action means), inaccurate performances (i.e. the bar in an incorrect penholder) and speed of action performances. As with the free execution task, action speed corresponded to the duration from when participants started to lift the hands from start position until the moment the bar was placed in a penholder.

A similar experimental set-up was used as described for the execution task. One exception was the use of a mono-colour bar instead of a bicolour bar. Participants watched a video clip in which the demonstrator (e.g. female adult, head masked) placed the mono-colour bar with or without application of the ESC principle in the left or in the right penholder (32 trials and 8 counterbalanced conditions). During each trial, participants were instructed to place the bar as fast as possible into the same penholder as that in which the demonstrator had placed it. They were allowed to start moving as soon as the demonstrator started moving. As with the free execution task, participants were unaware of task intentions.

Data registration

Synchronizations

An analog-digital converter (CED 1401 micro; Cambridge Electronic Design, UK) was used to synchronize visual cues presented on the monitor (using a multifunction DAQ, USB-6008 in combination with LabVIEW; National Instruments, US) with performances (button presses, sensors). Data were digitized at 500 Hz and stored on a personal computer (PC) for offline analysis. Signal software was used for data acquisition and extraction of the raw data.

To accurately calculate speed (in seconds) in both the free execution and imitation task, hand plates, support cradles and penholders were all connected to sensors: two sensors were positioned on the hand plate, one left and one right, for the left and right hand, respectively; two sensors were positioned on top of the bar support cradles; one sensor was positioned in the left penholder and one sensor in the right penholder. Signals from sensors were synchronized with the appearances of visual cues on the monitor by an analog-digital converter and signal software (version 4.03; Cambridge Electronic Design, UK). To measure efficiency (i.e. application of the ESC principle during the free execution task) and copying behaviour (i.e. means-end imitation, emulation and inaccurate performances during the imitation task), participants’ hand movements were continuously video-taped using an external video-camera (Canon Legria HRF 16), which was positioned behind the monitor screen and faced the table with task equipment.

Data analyses

The attention task

Accuracy level (i.e. correct button presses) and average reaction times (in seconds) were calculated across trials for all participants. Furthermore, for each participant, a performance ratio was calculated from the average accuracy level and reaction time by dividing the participants’ average percentage accuracy by the average time between two successive key presses (Cuypers et al., 2013). The independent variable was group (ASD vs neurotypical).

The free execution task

Accuracy (i.e. whether the bar was in the correct penholder with the correct end pointing up), efficiency (i.e. application of the ESC principle) and speed (in seconds) were calculated across trials. Applications of the ESC principle corresponded to thumb-up placements of the bar in the penholder. The total speed (e.g. Speed_total) was further divided into three separate components. They were defined respectively as the time from the stimulus onset to the start of the action, in particular to lifting the hand (Speed_start), the time from lifting the hand to lifting the bar (Speed_hand) and the time from lifting the bar until its placement in the penholder (Speed_bar). The independent variable was group (ASD vs neurotypical).

The imitation task

Accuracy (i.e. placing the bar in the correct penholder), the amount of means-end imitations, emulations, inaccurate performances and the speed (in seconds) were calculated across trials. Means-end imitation was analysed from video recordings and corresponded to copying of a demonstrated means-end performance that was applied either with or without the ESC principle. Emulation was also analysed from video recordings and corresponded to placing the bar into a similar penholder using another means than that demonstrated for the trial. Modelled ESC involved thumb-up placements of the bar in the penholder by a demonstrator on the video. Independent variables were group (ASD vs neurotypical) and modelled ESC (ESC vs no ESC).

Statistical analyses

The Shapiro–Wilk test was used to test normality of data. In cases of normal distribution of data (e.g. chronological age), group differences were calculated using an unpaired t-test. In cases where the assumption of normal distribution of data was not met (e.g. attention, imitation and free execution data), group differences were calculated using Mann–Whitney U tests (U).

Mixed linear regression models were used to analyse reaction times over different trials of free executions and imitations of actions in groups, and to calculate group differences. All analyses were performed using the statistical software SPSS (Version 24.0; SPSS Inc., Chicago, IL, USA) and JMP Pro (Version 12; SAS Institute Inc., Cary, NC, USA).

Results

The attention task

Overall accuracy level ranged from 83.3% to 100%. Mean accuracy rates for the two groups (20 ASD and 19 neurotypical adults) did not differ significantly (U = 133.5; Z = −1.7; p = 0.113; r = 0.27). In addition, groups did not differ on average reaction time (U = 235.0; Z = 1.3; p = 0.214; r = 0.21). Average reaction times of participants varied from 0.76 to 1.16 s. Groups also did not differ on their performance ratio (U = 132.5; Z = −1.6; p = 0.107; r = 0.26).

Subanalyses revealed that groups of college and university students (15 with ASD and 19 neurotypical) did not differ on measures of accuracy (U = 101.5; Z = −1.53; p = 0.157; r = 0.26), reaction time (U = 148.0; Z = 0.191; p = 0.864; r = 0.033) or performance ratio (U = 126.5; Z = −0.56; p = 0.584; r = 0.10).

Subanalyses further revealed that participants with an ASD who scored positive on the SRS (11 ASD) did not differ significantly on measures of accuracy (U = 63.5; Z = 1.10; p = 0.295; r = 0.25), reaction time (U = 57.0; Z = 0.57; p = 0.603; r = 0.13) or performance ratio (U = 47.5; Z = −0.15; p = 0.882; r = 0.03) from participants with an ASD who scored negative on the SRS (nine ASD).

The free execution task

All participants placed the bicolour bar identical to the way demonstrated on video in the correct position (i.e. in the correct penholder with the correct bar end pointing up; 100% accuracy in both groups). Participants with ASD were significantly less efficient in their performances than neurotypical adults, since they applied the ESC principle during free action executions less often (U = 115.5; Z = −2.1; p = 0.035; r = 0.34). Participants with ASD applied the ESC principle in 74.6% and neurotypical adults in 84.9% of their actions.

Results of a mixed linear regression model with group, trial and their interaction as fixed factors, and individual participants as random factor, further showed that there were no significant differences between participants with ASD and neurotypical participants for total speed of actions (Speed_total: F_{1, 37} = 1.74; p = 0.196), speed of hand lifting (Speed_start: F_{1, 37} = 1.16; p = 0.288) and speed of bar grasping (Speed_hand: F_{1, 37} = 3.26; p = 0.079). In contrast, participants with ASD were significantly slower than neurotypical adults in the bar placement stage of actions (Speed_bar: F_{1, 37} = 5.62; p = 0.023).

Subanalyses of performances of both groups of college and university students (15 with ASD and 19 neurotypical) showed a similar pattern: both groups tended to differ on ESC applications (U = 89.5; Z = −1.85; p = 0.066; r = 0.32), but not on reaction times (Speed_start: F_{1, 32} = 0.98; p = 0.329; Speed_hand: F_{1, 32} = 1.88; p = 0.180; Speed_total: F_{1, 32} = 1.01; p = 0.322), with the exception of bar placements in the penholder (Speed_bar: F_{1, 32} = 4.21; p = 0.048), which participants with ASD performed slower.

In addition, both SRS subgroups of participants with ASD (11 scoring positive and 9 scoring negative on the SRS) applied the ESC principle to a similar extent (U = 60.0; Z = 0.80; p = 0.456; r = 0.18). Moreover, speed of actions did not differ between these subgroups (Speed_start: F_{1, 18} = 2.14; p = 0.161; Speed_hand: F_{1, 18} = 0.10; p = 0.752; Speed_bar: F_{1, 18} = 0.02; p = 0.898; Speed_total: F_{1, 18} = 0.22; p = 0.643).

The imitation task

Participants always placed the mono-colour bar in the same penholder as the demonstrator (100% accuracy in both groups). The two groups (n = 20 ASD and n = 19 neurotypical) did not differ on the amount of means-end imitations (U = 161.0; Z = −0.82; p = 0.428; r = 0.13) and emulations (U = 219.0; Z = 0.82; p = 0.428; r = 0.13). Actions were imitated according to means-end in around half of all cases (ASD: in 47.7% of trials; neurotypical adults: in 53.8% of trials).

Results of a mixed linear regression model with group, trial and their interaction as fixed factor, and individual participants as random factor, showed that there was no difference between participants with ASD and neurotypical participants on total speed of imitated actions according to means-end (Speed_total: F_{1, 37} = 0.55; p = 0.461), or any of its subcomponents (Speed_start: F_{1, 37} = 0.12; p = 0.729; Speed_hand: F_{1, 37} = 0.06; p = 0.800; and Speed_bar: F_{1, 37} = 1.18; p = 0.285).

Subanalyses of the 16 trials in which the demonstrator applied the ESC principle revealed that the two groups did not differ on the amount of means-end imitations (U = 185.5; Z = −0.13; p = 0.901; r = 0.02) and emulations (U = 194.5; Z = 0.13; p = 0.901; r = 0.02). Actions were imitated according to means-end in around three quarters (74.4%) of all cases (ASD: in 73.8% of trials; neurotypical adults: in 75.0% of trials). Note that the remaining actions were imitated according to emulation. Subanalyses of the 16 trials in which the demonstrator did not apply the ESC principle revealed that the two groups did not differ on the amount of means-end imitations (U = 155.5; Z = −1.04; p = 0.336; r = 0.17) and emulations (U = 224.5; Z = 1.04; p = 0.336; r = 0.17). Actions were imitated according to means-end in around a quarter (26.9%) of all cases (ASD: in 21.6% of trials; neurotypical adults: in 32.6% of trials); the remaining actions were imitated according to emulation.

Subanalyses of groups of college and university students (15 with ASD and 19 neurotypical) showed a similar pattern: groups did not differ from each other on means-end imitation (U = 112.5; Z = −1.04; p = 0.302; r = 0.18), emulation (U = 172.5; Z = 1.04; p = 0.302; r = 0.18) or reaction times (Speed_start: F_{1, 33} = 0.21; p = 0.652; Speed_hand: F_{1, 33} = 0.07; p = 0.799; Speed_bar: F_{1, 33} = 0.52; p = 0.477; Speed_total: F_{1, 33} ⩽ 0.01; p = 0.960).

Furthermore, both SRS subgroups of participants with ASD (11 scoring positive and 9 scoring negative on the SRS) did not differ in the amount of means-end imitations (U = 46.0; Z = −0.27; p = 0.824; r = 0.06) and emulations (U = 53.0; Z = 0.27; p = 0.824; r = 0.06). Finally, speed of imitations also did not differ between these subgroups (Speed_start: F_{1, 18} = 0.95; p = 0.343; Speed_hand: F_{1, 18} = 1.14; p = 0.299; Speed_bar: F_{1, 18} = 0.03; p = 0.859; Speed_total: F_{1, 18} = 1.02; p = 0.325).

Discussion

This study is the first to systematically investigate the imitation aptitudes of intellectually able young adults with ASD compared to age-matched neurotypical adults, by analysing the extent to which they copy the means-end structure of demonstrated actions requiring anticipatory motor planning abilities (i.e. the application of ESC principle) using Rosenbaum et al.’s (1990) bar-transport task. Participants’ copying behaviour with respect to the demonstrators’ action means (i.e. with vs without ESC) and action result (i.e. placements of a mono-colour bar in the left vs right penholder) was analysed. In addition, their action speed was investigated. Results revealed that young adults with ASD imitated the demonstrator’s actions (i.e. adopting their means-end structure) as quickly and often as neurotypical young adults. In addition, they were as accurate in their performances as their neurotypical peers. In contrast, results revealed anticipatory motor planning problems in ASD, while participants with ASD compared to neurotypical adults showed reduced application of the ESC principle during free action executions. Moreover, adults with ASD were slower than neurotypical adults in fine-tuning placements of the bar in the penholders during free action executions. Although participants with ASD applied the ESC principle less frequently than neurotypical adults, groups showed a similar level of accuracy (i.e. placing the bar in the correct penholder with the correct bar end up).

There are several possible explanations for our finding that young adults with ASD showed preserved imitation skills. One is that these intellectually able individuals with ASD outgrew their imitation problems (Biscaldi et al., 2014). A second explanation is that these individuals always had motor planning rather than imitation problems. Longitudinal studies using well-controlled experimental methodologies are needed to investigate developmental pathways. An alternative explanation is that the imitation task wasn’t sensitive enough to detect impairments in intellectually able adults. To the best of our knowledge, only one other study has investigated means-end imitation skills in this population relative to age-matched neurotypical adults (Leighton et al., 2008). Leighton et al.’s pen-and-cups imitation task was more demanding than our imitation task. In order to minimize imitation errors, participants in Leighton et al.’s study needed to keep track of three dimensions of actions: object, effector and ESC. In each trial, participants saw a demonstrator moving a centrally located pen into one of two coloured cups (object) with his right or his left hand (effector), while grasping the pen with his thumb pointing up or down (ESC). Results of Leighton et al.’s study revealed impaired performance by adults with ASD on the imitative version of the pen-and-cups task. The same participants then completed two non-imitative versions of the task. The geometric version of the task required participants to perform actions specified by the movement of abstract geometric shapes. The verbal version of the task required participants to describe the observed actions. Adults with ASD were as impaired on each non-imitative version of the task as they were on the imitative version, suggesting that the impaired performance on the imitation task was not due to an imitation deficit per se. Instead, the authors concluded that more general factors, such as underlying difficulties with rapid shifting of attention between action components, contributed to the poor performance on their imitation task. The bar-transport imitation paradigm used in this study controlled for attention problems, but may have not been demanding enough for intellectually able adults because they were free to imitate actions with their hand of choice (i.e. left or right). Hence, they only needed to keep track of two dimensions of the action: the object and ESC (i.e. not the effector). A fourth plausible explanation is that the imitation abilities of neurotypical adults were underestimated in this study. Neurotypical adults who often applied the ESC principle during the free execution task could have been confused by observations of inefficient actions, and therefore could have shown a reduced tendency to reproduce actions that were relatively inefficient to reach goals. However, subanalyses revealed that both groups imitated inefficient actions less often (i.e. 73.1% emulated and 26.9% imitated) than efficient ones (i.e. 74.4% imitated and 25.6% emulated), and group differences were not significant. A final reason why we unexpectedly observed preserved imitation skills in ASD might be related to the fact that some of intellectually able adults are situated at the end of the ASD spectrum (i.e. have less severe ASD symptoms) and may therefore not represent the full range of deficits observed across the autism spectrum. However, it is important to note that all participants in the ASD group had undergone a large battery of widely used and validated diagnostic tests carried out by a multidisciplinary team, although for most of these participants, their diagnosis hadn’t been confirmed recently (i.e. the mean age of last confirmation of ASD diagnosis was 11.8 years). We implemented the SRS-A (Noens et al., 2012) merely to exclude neurotypical participants possibly fulfilling ASD criteria, and not as a diagnostic instrument to validate ASD diagnosis. However, SRS-A data from participants with ASD revealed that nine of them scored negative on SRS-A. One possibility is that these participants might have failed to report the truth with regard to their social difficulties due to a lack of insight in their own social (dis)abilities. Alternatively, they compared their current autistic traits to their autistic traits throughout childhood, which might have been considerably worse, and, as a result, they might have failed to report current ASD symptoms. To ensure that these participants formed a representative ASD sample, both subgroups of participants scoring negative and positive on SRS-A were compared on all parameters in the experimental and control tasks. Results showed that there was no difference between subgroups, indicating that the selected ASD sample represented a relatively homogeneous sample of intellectually able young adults with ASD.

One striking finding in this study was the persistence of anticipatory motor planning problems in intellectually able adults with ASD. Results of our study revealed that young adults with ASD relative to age-matched neurotypical adults applied the ESC principle less frequently during the free execution task, even though both groups showed a similar degree of accuracy (i.e. placing the correct bar end up in the correct penholder as shown in the video clip). In other words, participants with ASD less frequently chose to grasp the bar in an awkward fashion to ensure a comfortable hand posture at the end of the movement. By consequence, they were forced to put the bar in the penholder with a thumb-down hand position (i.e. an uncomfortable end position), which was more demanding and required more time (i.e. Speed_bar: p = 0.023). Other evidence to suggest anticipatory motor planning problems in the free execution task is that the speed from lifting the hands until picking up the bar approached significance (ASD participants were slower than neurotypicals).

One explanation is that weak anticipatory motor planning is a persistent deficit in ASD. Indeed, anticipatory motor planning problems in ASD have been reported in many studies (e.g. Gonzalez et al., 2013; Gowen and Hamilton, 2013; Hughes, 1996). For instance, in a joint-action paradigm in which participants had to give a tool to a confederate either to be used or placed down, adults with ASD compared to neurotypicals had a more variable performance and showed fewer appropriate responses with regards to both goals (Gonzalez et al., 2013). An alternative explanation is that reduced application of ESC in young adults with ASD might result from intellectual functioning differences between both groups. It is important to note that all participants followed mainstream education, thus reducing the likelihood that intellectual functioning differences contributed to the findings. We addressed this quite unlikely possibility, however, by excluding five high-school students with ASD from our data analyses. Results of the age-matched college and university students (15 ASD and 19 neurotypical ones) revealed identical trends as in the broader sample, suggesting that our finding of anticipatory motor planning problems in ASD is unlikely to reflect differences in intellectual functioning. However, given that we put limited emphasis on intellectual functioning in our study, we recommend future research addressing our topic to additionally explicitly focus on intellectual functioning. A final explanation could be that group differences in the free execution task are attributable to differences in attention. Accuracy and performance ratio measures in the attention task were approaching significance. Importantly, accuracy measures were at ceiling, suggesting a true significant effect failed to come out. Hence, adults with ASD might have experienced attention problems during task performances. However, participants with ASD were able to perform all actions with 100% accuracy in the free execution task, in which similar type of stimuli were used as in the attention task. Moreover, in case observed group differences were to be attributed to attention problems only, group differences in imitation should also have been found.

The results of this study must be considered within the limitations of its design and sample. First of all, there was a difference in gender ratio between the groups. Nevertheless, our ASD sample resembled the suggested male–female ASD ratio of boys with ASD outnumbering girls with ASD with a ratio of 4:1 (Lai et al., 2014). Furthermore, we did not administer IQ measures. However, all participants followed mainstream high schools, college or university. We conducted additional analyses for a subgroup excluding participants following high-school education, and were able to replicate the results of the larger sample, although the results were underpowered.

These findings could potentially have implications for clinical practice. Intellectually young adults with ASD might benefit from imitation training in which they can learn relevant motor planning skills needed during daily life (e.g. grasping a cup out of a cabinet; using a screwdriver or other manual tools) (see Brown et al. (2008) for a comprehensive literature review on imitation training in ASD).

To conclude, intellectually able young adults with ASD are impaired in anticipatory motor planning as indicated by reduced application of the ESC principle during free action executions, as well as lower speed at the end of these actions. However, we found no evidence for impairments in copying the means-end structure of actions. Longitudinal studies are needed to investigate trajectories of imitation and motor planning difficulties in children and adults with ASD across the life span. Nevertheless, observing actions of others might serve as a template for more efficient action planning in young adults with ASD by helping them to acquire relevant motor skills.

Supplemental Material

AUT698454_Lay_Abstract – Supplemental material for Preserved imitation in contrast to limited free application of comfortable hand actions in intellectually able young adults with an autism spectrum disorder

Supplemental material, AUT698454_Lay_Abstract for Preserved imitation in contrast to limited free application of comfortable hand actions in intellectually able young adults with an autism spectrum disorder by Caroline Beelen, Koen Cuypers, Lise van Schuerbeeck, Marijke Braeken, Veerle Ross, Ellen Jongen, Raf Meesen and Marleen Vanvuchelen in Autism

Footnotes

Acknowledgements

We would like to thank all individuals who participated in this study. Furthermore, we would like to thank Dr Francesca Solmi (Centre for Statistics of Hasselt University) and Dr Wolfgang Scharke (University Hospital of RWTH Aachen University) for assistance with data analyses, and Dr Liz O’Nions for assistance in adjusting the paper in a way that improved our manuscript significantly.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Ethical approval

This study was approved by the ethics committee of the Hasselt University and University Hospital Louvain (Belgium) before data collection (B322201421678, S56913) and confirmed to the principles stated in the Declaration of Helsinki, 1964. All participants gave informed consent prior to participation.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was partially supported by a Foundation Marguerite-Marie Delacroix grant (Tienen, Belgium) given to M.B. and V.R.

References

American Psychiatric Association (APA) (1994) Diagnostic and Statistical Manual of Mental Disorders (DSM IV). 4th ed. Washington, DC: APA.

American Psychiatric Association (APA) (2000) Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR). 4th ed. Text revision. Washington, DC: APA.

Biscaldi

Rauh

Irion

. (2014) Deficits in motor abilities and developmental fractionation of imitation performance in high-functioning autism spectrum disorders. European Child & Adolescent Psychiatry 23(7): 599–610.

Brown

Poulson

(2008) Discriminating which fork to use: Teaching selective imitation to people with autism. Research in Autism Spectrum Disorders 2: 199–208.

Constantino

Davis

Todd

. (2003) Validation of a brief quantitative measure of autistic traits: Comparison of the social responsiveness scale with the autism diagnostic interview-revised. Journal of Autism and Developmental Disorders 33(4): 427–433.

Cuypers

Leenus

van den Berg

. (2013) Is motor learning mediated by tDCS intensity? PLoS ONE 8(6): e67344.

Edwards

(2014) A meta-analysis of imitation abilities in individuals with autism spectrum disorders. Autism Research 7(3): 363–380.

Gonzalez

Glazebrook

Studenka

. (2013) Motor interactions with another person: Do individuals with autism spectrum disorder plan ahead? Frontiers in Integrative Neuroscience 7: 23.

Gowen

Hamilton

(2013) Motor abilities in autism: a review using a computational context. Journal of Autism and Developmental Disorders 43(2): 323–344.

10.

Hobson

Lee

(1999) Imitation and identification in autism. Journal of Child Psychology and Psychiatry 40(4): 649–659.

11.

Hughes

(1996) Brief report: planning problems in autism at the level of motor control. Journal of Autism and Developmental Disorders 26(1): 99–107.

12.

Jiménez

Lorda

Méndez

(2014) Emulation and mimicry in school students with typical development and with high functioning autism. Journal of Autism and Developmental Disorders 44: 1597–1608.

13.

Lai

M-C

Lombardo

Baron-Cohen

(2014) Autism. The Lancet 383: 896–910.

14.

Leighton

Bird

Charman

. (2008) Weak imitative performance is not due to a functional ‘mirroring’ deficit in adults with autism spectrum disorders. Neuropsychologia 46: 1041–1049.

15.

Noens

De la Marche

Scholte

(2012) SRS-A-Screeningslijst voor autismespectrumstoornissen: Handleiding. Amsterdam: Hogrefe.

16.

Rogers

Pennington

(1991) A theoretical approach to the deficits in infantile autism. Development and Psychopathology 3: 137–162.

17.

Rosenbaum

Marchak

Barnes

. (1990) Constraints for action selection: overhand versus underhand grips. In: Jeannerod

(ed.) Attention and Performance XIII: Motor Representation and Control. London: Psychology Press, pp.321–342.

18.

Stoit

van Schie

Slaats-Willemse

. (2013) Grasping motor impairments in autism: Not action planning but movement execution is deficient. Journal of Autism and Developmental Disorders 43(12): 2793–2806.

19.

Tomasello

Carpenter

Call

. (2005) Understanding and sharing intentions: The origins of cultural cognition. Behavioral Brain Science 28(5): 675–691.

20.

Tomasello

Savage-Rumbaugh

Kruger

(1993) Imitative learning of actions on objects by children, chimpanzees, and enculturated chimpanzees. Child Development 64(6): 1688–1705.

21.

Vanvuchelen

Roeyers

De Weerdt

(2011) Do imitation problems reflect a core characteristic in autism?: Evidence from a literature review. Research in Autism Spectrum Disorders 5(1): 89–95.

22.

Wunsch

Henning

Aschersleben

. (2013) A systematic review of the end-state comfort effect in normally developing children and in children with developmental disorders. Journal of Motor Learning and Development 1(3): 59–76.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.44 MB