Anxiety in Asperger’s syndrome: Assessment in real time

Abstract

Anxiety is a major problem for many people with Asperger’s syndrome who may have qualitatively different fears from a non-Asperger’s syndrome population. Research has relied on measures developed for non-Asperger’s syndrome populations that require reporting past experiences of anxiety, which may confound assessment in people with Asperger’s syndrome due to problems with autobiographical memory as are often reported in this group.

Experience sampling methodology was used to record real-time everyday experiences in 20 adults with Asperger’s syndrome and 20 neurotypical adults. Within-subject analysis was used to explore the phenomenology of thoughts occurring in people with Asperger’s syndrome when they were anxious. Comparisons were made with the group that did not have Asperger’s syndrome. The Asperger’s syndrome group were significantly more anxious than the comparison group. Factors associated with feelings of anxiety in the Asperger’s syndrome group were high levels of self-focus, worries about everyday events and periods of rumination lasting over 10 min. People in the Asperger’s syndrome group also had a tendency to think in the image form, but this was not associated with feelings of anxiety. The results are discussed with reference to psychological models of Asperger’s syndrome, cognitive models of anxiety and implications for psychological therapy for this group.

Keywords

anxiety Asperger’s syndrome cognitive processes experience sampling methodology

Introduction

Transition to adulthood is challenging for many people with Asperger’s syndrome (AS) as social impairments continue into adulthood affecting interpersonal relationships and social networks (Lawrence et al., 2010), leading to social isolation (Cederlund et al., 2008), restricted lives (Jennes-Coussens et al., 2006) and problems in living independently (Engstrom et al., 2003; Lawrence et al., 2010). There is little guidance for the support and management of AS in adulthood (Department of Health, 2010), but positive outcomes are associated with timely diagnosis, treatment of co-morbidity and appropriate and accessible support (Kim et al., 2000; Renty and Roeyers, 2006). Adults with AS are vulnerable to co-morbid mental health difficulties, particularly anxiety disorders (Hofvander et al., 2009; Skokauskas and Gallagher, 2010; Van Steensel et al., 2011), which may be inversely related to the severity of autistic symptomatology (e.g. Sterling et al., 2008; White et al., 2009). Personal accounts also emphasise anxiety and stress as everyday experiences for people with AS (e.g. Dubin, 2009).

Stress is subjectively experienced when the perceived demands of an individual’s situation exceed the perceived available resources and some models of stress emphasise minor daily hassles over major life events in the development of stress, which appears to be the case for people with AS (Sze and Wood, 2007), who seem particularly susceptible to experiencing stress (Baron et al., 2006a) with daily lives characterised by unpredictability (Dubin, 2009; Grandin, 2006). Appraisal of stress, experience of anxiety and sources of stress appear to be different for people with AS (Baron et al., 2006b; Dubin, 2009; Gaus, 2011; Gillott and Standen, 2007), leading to hyper-arousal and further stress. Wood and Gadow (2010) in their conceptual review of anxiety in autism spectrum disorder (ASD) identify that anxiety may be a consequence of ASD, a moderator of the core ASD symptomatology or a proxy for the core symptoms, while noting that the potential for reports of anxiety in ASD to be of low validity. Similarly, the relationship between state anxiety and trait anxiety (Endler and Kocovski, 2001) has still to be fully determined in AS.

To date, despite research and clinical reports indicating high levels of anxiety in this group (see White et al. (2009) and Wood and Gadow (2010) for review of anxiety in ASD), there is very little research detailing the content, frequency and severity of anxiety required to develop more appropriate clinical interventions for people with AS (Hare, 2013). One problem with measures developed for neurotypical (NT) populations is the requirement for retrospective description of anxious thoughts and feelings, given that people with AS may show impaired autonoetic functioning (Bowler et al., 2000) and/or a degree of alexithymia (impaired ability to identify and describe feelings) is associated with AS (Berthoz and Hill, 2005). Therefore, the phenomenology of anxiety in people with AS should be investigated rather than assumed, as idiosyncratic explanations and/or qualitatively different experiences of anxiety and related thoughts and feelings may not be captured by existing tools.

One method of capturing real-time cognitive and behavioural events is experience sampling methodology (ESM) (Palmier-Claus et al., 2011) that allows for the gathering of perceptions, thoughts and feelings shortly after they occur and reduces the effects of memory on recording experience (Hektner et al., 2006). ESM reveals how much time a person spends thinking or doing something and provides a relatively undistorted account of daily life. There are three main forms of ESM (Reis and Gable, 2000): interval-contingent sampling (participants self-report after a designated interval for a pre-set amount of time), event-contingent sampling (participants self-report when a pre-designated event occurs) and signal-contingent sampling (participants self-report when prompted by a randomly timed signal), the last one being advantageous in that it allows for the sampling of a representative schedule of times and avoids any expectancy effects that may come from having prior knowledge of the sampling period.

Hurlburt et al. (1994) used a similar method, descriptive experience sampling, to investigate the everyday experience of three adults with AS. This involved use of a beeper sounding at random intervals to signal subjects to make brief notes about current mental contents for later interview discussion. Hurlburt and colleagues found that if inner experience was reported at all, it involved only visual images (inner speech, feelings, etc. were never reported) and proposed that the tendency to have an inner experience populated by images may explain difficulties with mentalisation, suggesting that ‘it is extremely difficult, if not impossible, to represent another’s point of view using only visual images’ (Hurlburt et al., 1994). A recent study (Hintzen et al., 2010) used paper-based ESM to explore the social needs and interactions in daily life of eight adults with ASD compared to a matched control group over a 6-day period and indicated no group differences in time alone and levels of social activity, but the ASD group reported more time with familiar people and more negative affect and anxiety when with unfamiliar people.

Given the characteristics common to people with AS and suggestions highlighting the importance of coping with everyday life, in everyday contexts (Gaus, 2011), the personal data assistant (PDA) is particularly well suited. PDAs are ideal in that they can accommodate suggested modifications for working therapeutically with individuals with AS (e.g. Moree and Davis, 2010). PDAs are primarily visual and allow individuals to have control to work at their own pace. PDAs allow for repetition and the opportunity for individuals to build up coping skills within the everyday environment, which is particularly important when considering the consolidation of principles via computer-delivered interventions (Clough and Casey, 2011). They can function as a ‘coach’, providing prompts in real time (i.e. throughout the day and across environments), strategies which have been suggested to aid the generalisation of skills (e.g. Cardaciotto and Herbert, 2004; Hare, 1997; Weiss and Lunsky, 2010). PDAs may be more suitable as they remove the social demands of therapeutic work and reduce reliance on memory, which may be problematic, given the difficulties with autobiographical memory common among individuals with AS. Furthermore, programmes can be designed in such a way that techniques can be reduced to step-by-step directions that is applicable across situations and provide a structured and consistent approach.

Computerised interventions offer the opportunity to capture real-life, moment-to-moment experiences. Baron et al. (2006a) note that common interventions for individuals with autism spectrum conditions (ASCs) often stress the importance of the use of external strategies (i.e. involving other people, changing the environment), with little emphasis on self-management, which may be helpful given the constant interplay with the environment. They also stress the importance of repeated assessment and observations in various settings when investigating stress and coping in this population as it allows for examination of temporal change and the idiosyncratic interaction between an individual and their environment. Computerised ESM (i.e. PDA) provides the opportunity for this type of exploration as it is well suited for the investigation of fluctuating stress and anxiety. Furthermore, it allows for the application and evaluation (i.e. pre- and post-subjective ratings) of psychotherapeutic input on a moment-to-moment basis.

Previous studies have used computerised interventions for the treatment of anxiety in the general population, and PDAs have been used as adjuncts to interventions targeting independence for adolescents with ASDs (e.g. Myles et al., 2007). Momentary assessment has been used to explore the experiences and social lives of adults with AS (e.g. Hintzen et al., 2010), but no study to date has used a PDA to investigate stress and state anxiety in adults with AS using ESM.

This article reports on an exploratory study with the aim of exploring the experiences of state anxiety and everyday stress in individuals with AS compared to a NT control group by examining the phenomenology of thoughts, attentional styles, thought content and situations that are associated with anxiety in people with AS and also by examining whether such presumed state anxiety is the result of rumination or from an interruption to rumination resulting in increased state anxiety

Method

Ethical approval for the study was obtained from the National Autistic Society (NAS) and The University of Manchester Ethics Committee.

Design

A mixed within- and between-groups was employed to allow for detailed investigation within the AS group in regard to factors associated with anxiety and for comparisons of these factors with a NT group. Both groups completed a questionnaire developed for use with a Palm Pilot together with measures of anxiety and depression. IQ scores and verbal comprehension were used to determine eligibility. A power calculation indicated that 24 subjects in each group were required for identification of significant predictors of anxiety with a maximum of five predictors identified, which was derived using the 5:1 rule that utilises the minimum R-squared that can be found statistically significant at the p < 0.01 level with a power of 0.80 for this number of independent variables and sample size (see Hair et al., 1995).

Inclusion criteria for the AS group were (1) a verifiable diagnosis of AS received from a medical practitioner or psychologist, (2) over 18 years and below 65 years old, (3) no known acquired head injury and (4) fluent in English (Flesh–Kincaid grade level of 8.8/reading age of 9.4 years) and (5) full-scale IQ (FSIQ) score >70. The AS participants were recruited via (1) an advertisement on the UK NAS website (N = 6), (2) AS support groups in Yorkshire (N = 5) and Merseyside (N = 7) and (3) ‘snowball’ recruitment by existing participants of other people with AS (N = 10). In total, 48 participants were recruited (28 AS participants, 20 control participants), but technical difficulties resulted in the loss of eight sets of data from the AS group. The principal researcher (C.W.) met with each participant on three occasions and total contact time for each participant was around 3 h.

Measures

Palm Pilot Questionnaire – a 13-item questionnaire (Appendix 1) was developed for use with a Palm Pilot (OS R version 3.1 or above) and the ESM software by the principal researcher (C.W.). Questions 1, 2, 7, 8, 9 and 13 were adapted from an ESM study of anxiety in a student population (deVries et al., 2001), and Question 3 was adapted from Hurlburt et al. (1994), with the remaining questions developed for this study. The use of free source code from the National Science Foundation (http://www.experience-sampling.org) allowed various response types, including ‘text’ (100 characters max.), ‘multiple choice’ and ‘visual analogue’ questions. Text answers were coded as ‘external coordinates of experience’ (what the person was doing, how many people they were with) and ‘internal co-ordinates of experience’ (internal or external focus of thought, whether the thought including other people and whether the thought was anxious). Coding was undertaken by the principal researcher (C.W.) and a trainee clinical psychologist (S.W.) and inter-rater reliability analysis indicated acceptable kappa statistics of 0.72–0.96.

The Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983). This is a 14-item measure of anxious and depressed feelings rated on a 4-point scale with a cut-off score of 8 for both sub-scales taken as indicating clinical anxiety or depression. It has been used in previous research with people with AS (Abell and Hare, 2005).

Wechsler Abbreviated Scale of Intelligence (WASI) (Wechsler, 1999). A measure of general intellectual ability, comprising four subtests (Vocabulary, Similarities, Block Design and Matrix Reasoning), yielding Verbal IQ (VIQ), Performance IQ (PIQ) and a FSIQ-4 scores.

British Picture Vocabulary Scale II (BPVS II) (Dunn et al., 1982). This was used to assess receptive verbal ability in Standard English.

Procedure

Participants were given a pre-programmed Palm Pilot to carry with them during the day for 3 days. This was pre-programmed to beep randomly 10 times between 10:00 and 16:00 h. Upon hearing the beep, and if safe to do so, participants responded to the questionnaire items within 300 s and had 300 s to complete each question, otherwise the Palm Pilot automatically turned off until the next activation. This was to minimise the possibility of memory bias influencing the experiences reported, with the Palm Pilots being locked in between activation to ensure the privacy and confidentiality of subjects’ responses. Data were downloaded to a PC using standard Palm Pilot to PC data synchronisation (iESP software and the Palm Pilot ‘Hot Sync’ function) as a text file, converted into an Excel spreadsheet and exported into SPSS version 16.

Results

A total of 20 AS and 20 NT participants took part in the study. Of the AS participants 14 (70%) were male and 6 (30%) were female. Mean age of the AS group was 32.1 years (SD = 12.17; range = 18–59 years) and that of the NT group 34.7 years (SD = 11.94; range = 23–61 years). No participants reported any current psychiatric diagnoses. One AS participant had been referred for paranoia before receiving a diagnosis of AS, and other AS participants reported anxiety problems but had not sought professional help. All of the datasets were substantially complete with only 0.36% missing data and thus satisfied the criterion of all participants having completed at the least 33% of ESM data points. Age, BPVS II and HADS scores were not normally distributed. Lognormal transformation was attempted but did not normalise the distributions, and non-parametric statistics were subsequently used with two-tailed probabilities reported. No significant differences were found between the two groups except for HADS anxiety (U (20, 20) = 3, p = < 0.001) and HADS depression (U (20, 20) = 64, p = < 0.001) scores, with mean HADS anxiety score for the AS group of 13.1, indicative of clinical anxiety (Table 1).

Table 1.

Sample characteristics.

	AS group	NT group
Group size (N)	20	20
Age	Mean: 32.15	Mean: 34.7
	SD: 12.17	SD: 11.94
	Range: 18–59 years	Range: 23–61 years
Gender	Male: 14	Male: 9
Gender	Female: 6	Female: 11
WASI score	Mean: 113.65	Mean: 116.40
	SD: 10.43	SD: 11.28
	Range: 91–128	Range: 102–135
BPVS II score	Mean: 16.47	Mean: 16.49
	SD: 0.93	SD: 0.58
	Range: 13.1–17 years	Range: 15.5–17 years
HADS anxiety score	Mean: 13.1	Mean: 3.3
	SD: 3.75	SD: 6.01
	Range: 7–19	Range: 0–8
HADS depression score	Mean: 5.75	Mean: 1.55
	SD: 3.6	SD: 1.28
	Range: 0–13	Range: 0–4

SD: standard deviation; AS: Asperger’s syndrome; NT: neurotypical; HADS: The Hospital Anxiety and Depression Scale; BPVS II: British Picture Vocabulary Scale II; WASI: Wechsler Abbreviated Scale of Intelligence.

In total, 30 data entries were collected for each participant resulting in 720 individual lines of data for each group. The first five sets of data for each participant were discarded as the pilot study showed that five or six periods of data entry were required in order to become familiar with the procedure. Other data were removed as unsuitable for analysis (e.g. random text). At least 33% of each person’s experience sampling activations had to be complete for data to be valid and included in the data analyses (Myin-Germeys et al., 2011). The final database contained 366 lines of data in the AS group and 285 lines in the NT group. Table 2 shows the comparative percentages of the phenomenology of AS thoughts and NT. Thoughts were categorised by focus, content and subjective appraisal: (1) Internal/external (i.e. percentage of thoughts were coded as being self-focused, for example, ‘I was thinking about a pain in my leg’ and external focused, for example, ‘I was thinking about the game I was playing’), (2) Did the thought involve another person? (i.e. whether the content of the thought concerned another person or not), (3) anxious thought (i.e. percentage of thoughts coded as anxious), (4) thought type (i.e. how the participant appraised the thought, for example, pleasurable, worrying, neutral) and (5) form of thought (i.e. how the participant described their thought, for example, like an image, like someone was talking to me). The AS group reported significantly more self-focused thoughts (U (20, 20) = 65, p = < 0.001), more confusing thoughts (U (20, 20) = 73, p = < 0.001) and worrying thoughts (U (20, 20) = 75, p = < 0.001). The AS group reported significantly more thoughts as images (U (20, 20) = 45.5, p = < 0.001), while the NT group reported more thoughts if they were talking to themselves (U (20, 20) = 75.5, p = 0.01). However, this latter difference appeared to be due to outlying data relating to a single participant and disappeared on exclusion of said participant from this analysis.

Table 2.

Characteristics of thoughts reported by AS and NT groups.

	AS group (%)	NT group (%)
1. Internal/external
Self-focus	34	4.6
External focus	66	95.4
2. Did the thought involve another person?
Yes	24.6	23.6
No	75.4	76.4
3. Anxious thought
Yes	16.3	6.3
No	83.6	93.7
4. Thought type
Neutral	21.4	15.1
Practical	17.8	36.2
Comforting	9.3	6.5
Pleasurable	12.1	24.4
Total: non-distressing	60.6	82.2
Worrying	15.6	3.6
Confusing	14	3.2
Annoying	5.5	6.8
Frustrating	1.4	7
Total: distressing	36.5	20.6
Other	2.7	3.6
Form of thought
Like an image	40	19.1
Like I was talking to myself	37.5	68.4
Like someone was talking to me	6.3	.7
Like written text	4.1	3.5
Like written symbols	1.1	1.1
Other	10.7	7.1

Table 3 shows comparative percentages of the ‘typicality’ of thoughts and length of time dwelling on a particular thought: (1) Had thought before? (i.e. whether the participant reported their thoughts as new or whether they had a similar thought previously). (2) How long? (i.e. how long the participant reported thinking about the thought they were asked to describe, e.g. reporting ‘I was thinking about how to start my essay’ and recording thinking about it for over 10 min). No significant difference was found for thoughts lasting between 5–10 min, but the AS group reported significantly more occasions of dwelling on a thought for over 10 min (U (20, 20) = 40, p = < 0.001).

Table 3.

Novelty and duration of thoughts reported in AS and NT groups.

	Asperger’s syndrome group (%)	Neurotypical group (%)
1. Had thought before?
Yes	81.6	83.8
No	8.8	13.7
Unsure	9.6	2.5
2. How long?
<min	14.8	38
1–5 min	32.1	49.3
5–10 min	22.2	8.1
Over 10 min	30.1	3.5
Unsure	0.8	1.1

AS: Asperger’s syndrome; NT: neurotypical.

Table 4 shows comparative percentages of the external factors occurring over the sampling period: (1) How many people with? (i.e. how many people participants reported as present when completing questionnaire). (2) Did the interruption to your thinking upset you? (i.e. whether request to complete the questionnaire and therefore interruption to their thought process had any effect on their emotions).

Table 4.

Context of reported thoughts.

	Asperger’s syndrome group (%)	Neurotypical group (%)
1 How many people with?
By myself	52.9	50.5
With one other person	16.2	19.4
With two or more people	31.0	30
2. Did the interruption to your thinking upset you?
No	78.4	88
It annoyed me	12.3	7.8
It made me feel nervous	4.7	0
I lost my train of thought	4.7	4.2

Participants rated their anxiety on a visual analogue scale (0–99). Anxiety scores for the AS group were skewed towards the more anxious end with NT group scores skewed in the other direction. Using a median split, scores of ≥60 were rated as anxious, with the AS group reported feeling anxious almost half of the time, compared to 7.3% for the NT group. There was no significant association between focus of thought and anxiety (χ² = 4.183 (Yates continuity correction) p = 0.041) with an odds ratio (OR) of 1.7851 (95% confidence interval (CI) 1.028–1.533) for feeling anxious when the focus was on the self rather than externally. There was no significant relationship between thoughts concerning other people and feelings of anxiety (χ² = 2.762). And 14% (N = 50) of all thoughts in the AS group were coded as anxious with a significant association between anxious thoughts and anxious feelings (χ² =31.972 (following Yates continuity correction)); p = < 0.01). The OR of an anxious feeling following an anxious thought was 4.017 (95% CI 2.094–7.706). However, when a person was experiencing a non-anxious thought, they were no more likely to be having a non-anxious feeling (OR = 0.517; 95% CI 0.439–0.609).

Of the reported anxious thoughts, 78.4% were of the Type 1 worry kind (Wells, 1997), for example, worry about a general non-cognitive event such as ‘I’m worried that the bus won’t be on time’; 15.7% were related to fear of negative evaluation by other people; and 5.9% were concerned about their health. There were no reports of Type 2 meta-worry (Wells, 1997) or intrusive thoughts.

There were eight possible ways of describing thought type with a significant association between reported thought type and feelings of anxiety (χ² =86.68; p = < 0.01) with 24% of the variation in frequency of anxious feelings attributable to thought type. The ORs for feeling anxious when worrying were compared to feeling anxious when having neutral thoughts. The ORs of feeling anxious when having a thought appraised as worrying or as confusing were 5.96 (95% CI 3.02–11.74) and 5.70 (95% CI 2.28–11.53). OR of feeling anxious when worrying was compared to feeling anxious when having confusing thoughts. Participants were no more likely to feel anxious whether the thought was appraised as worrying or confusing (OR = 0.95; 95% CI 0.38–2.37). Participants were no more likely to feel anxious whether the thought was appraised as frustrating or neutral (OR = 1.35; 95% CI 0.41–4.39). The form of thought refers to how the participant perceived their thought (e.g. as an image, as verbal dialogue). There was no significant association between focus of thought and anxiety (χ² = 11.076; p = 0.0530), but there was a significant association between novelty of thoughts and anxious feelings (χ² = 22.150; p = < 0.01).

Participants were also asked to estimate how long they had been thinking about the same thought with four options from less than 1 min to over 10 min. A significant association was observed between the lengths of time spent thinking a particular thought and anxious feelings (χ² = 34.94; p = < 0.01, degrees of freedom (DF) = 4). ORs were calculated for the ratio of feeling anxious when having a thought lasting 5–10 min compared to a thought lasting less than a minute (OR = 1.6; 95% CI 0.84–3.04), but this is non-significant as the lower bound interval is below 1. ORs were calculated for the ratio of feeling anxious when having a thought lasting over 10 min compared to a thought lasting less than a minute (OR = 4.44; 95% CI 2.28–8.649). Thus, when an AS participant was dwelling on a thought for over 10 min, there was an increased likelihood of reporting an anxious feeling. Although only periods of dwelling on a thought for over 10 min were significantly associated with anxiety, there was a linear increase in the number of occasions reported as anxious and rumination (Figure 1).

Figure 1.

Reported feelings of anxiety and rumination.

Three questions referred to events occurring around the participant, but there was no association between anxiety and what they were doing (χ² = 5.39) or with the alarm sounding (χ² = 5.39). Interruptions to thinking and feelings of anxiety was associated with anxiety (χ² = 10.03) but only at the p = 0.05 level, rather than the a priori 0.01 level.

The convergence of thoughts and feelings was calculated to determine whether AS participants were thinking about the task at hand or whether their thoughts were elsewhere; 52% of their thoughts were concerned with activity they were doing. A 2 × 2 χ² was carried out to discover whether there was a significant relationship between convergence of thought and feelings of anxiety. The χ value of 1.151 was not significant.

Discussion

The AS group reported significantly more worrying thoughts, more rumination and confusing thoughts, which were all associated with anxiety. They also reported significantly more thoughts as occurring as images, but this type of thought was not associated with feelings of anxiety. Finally, although anxious thoughts were significantly associated with feelings of anxiety, the majority of the reported feelings of anxiety occurred without the occurrence of an anxious thought in the AS group. Similarly, the AS group reported significantly more self-focused thoughts than the NT group, which were significantly associated with anxiety in the AS group. Of the reported thoughts, 14% were anxious, with 78.4% being Type 1 worry (i.e. worry about non-cognitive events; Wells, 1997), 15.7% fear of negative evaluation by other people and 5.9% health concerns. There were no reports of Type 2 meta-worry (Wells, 1997) or intrusive thoughts per se.

Rumination was significantly associated with feelings of anxiety in the AS group, who also engaged in this significantly more often than the NT group. No significant association was found between interruption to thoughts and feelings of anxiety, indicating that interruption to thinking does not result in feelings of anxiety in people with AS. Similarly, no particular situations were associated with anxiety as participants in both groups spent similar amounts of time in other peoples company, and this did not relate to feelings of anxiety in people with AS.

People with AS were significantly more anxious than people in the NT group (p = < 0.001) and were able to report their level of anxiety using a simple visual analogue scale that ranged from relaxed to nervous. It is important to note that the meaning of the word ‘nervous’ was checked with the participants before starting the study. Some of the AS participants would have used different words such as ‘stressed’ or ‘tense’, in place of nervous (e.g. they may have preferred the scale to range from relaxed to stressed). Nevertheless, they frequently reported high levels of nervousness and were significantly more nervous than the NT group. Only a small percentage of the thoughts occurring when they were feeling nervous were coded as anxious (14%).

These findings suggest that either anxious thoughts were often not occurring when the person was feeling anxious or that such thoughts were difficult to access and articulate. As the AS group reported significantly more confusing thoughts than the NT group, and these thoughts were associated with anxiety in the AS group, one explanation may relate to a relatively impoverished understanding of their own minds (Happé, 2003). On several occasions, AS participants reported feeling anxious, but the associated thought was a detailed description of the task they were doing (e.g. ‘I was thinking about a puzzle I was doing and whether the squares are in the same diameter’). One possible reason for this is that people with AS have difficulty recognising and appraising their own cognitions in relation to their feelings and instead reported them in relation to the task they were doing. This difficulty in expressing cognitions was not found in the NT group, who reported very few confusing thoughts and did not struggle to verbalise their internal world. This is congruent with the reports of alexithymia in people with AS (Fitzgerald and Bellgrove, 2006), with the high frequency of reported anxious feelings in the absence of any related cognitions being explained by a difficulty in appraising and verbalising such feelings. This difficulty was expressed by one participant who stated ‘There is no point in asking me about how I am feeling while I am feeling it: I do not know. I may know later’.

Thus, people with AS may be able to appraise thoughts as anxiety provoking, but are much less practised at discussing and articulating their thoughts. The results showed that when a person with AS did have a thought that was coded as anxious, they were more likely to also report feelings of anxiety, suggesting some degree of appraisal of their own cognitions. For example, one participant reported ‘I was panicking about the tasks I needed to complete today’ (coded as anxious and self-focused). The percentage of self-focused thoughts was significantly higher in the AS group than in the NT group, which were in turn associated with feelings of anxiety. Other research has reported increased private and public self-consciousness in people with AS (Abell and Hare, 2005).

It was noted that many participants with AS referred to the need for rules and structure in order to function, which may reflect more pervasive difficulties with aspects of executive functioning. For example, many of the worries reported by the AS group would not seem particularly anxiety-provoking to a non-anxious NT, which may suggest an apparent difficulty with flexible thinking in AS. One AS participant stated, ‘I work to the rule “if you don’t know what to do, do nothing”’. Such rule-governed behaviour might be the result of a difficulty in deviating from one ‘plan’ and/or modifying it in such a way as to create an internal representation of a new plan. The manifestation of repetitive or rule-governed behaviour may reflect dysexecutive problems as executive functioning facilitates shifting between different behaviours and resisting habitual but no longer useful behaviours. If such functioning is impaired, a person may either do nothing (as in the case of the participant above) or become anxious when required to do something for which they have no plan. As another AS participant stated, ‘If something I was expecting to happen one way, suddenly happens in a different way, it can throw me so much that I don’t respond. This also happens when I don’t know what is going on’.

This apparent difficulty coping with change may also reflect deficits in autobiographical memory functioning. Autobiographical memory comprises both personally experienced events (personal episodic memories) and self-related information (personal semantic memories) and contributes to solving social problems (Goddard et al., 2007), forming and maintaining social relationships and provides information for social communication and interaction. Several of the reported anxious thoughts concerned social and interpersonal expectations, for example, ‘I was thinking about how people communicate when somebody approaches them, do they communicate excuse me please’. This may be understandable in terms of poor autobiographical memory in that the person with AS may be unable to recall past episodes where they have been in similar situations. Autobiographical memory impairment in ASD appears to be due to a lack of bias for self-referential material, rather than problems in generating or encoding such material in the first instance, resulting in individuals with ASD lacking superiority of access to personal episodic memories for self-experienced events (Hare et al., 2007). In AS, self and emotion may be less important memory markers, making recall of autobiographical memory more difficult (Bowler et al., 2000).

Limitations

It is recognised that this study has a number of limitations, including the fact that the two groups were not matched for either state or trait anxiety prior to the study and that no provision was made for examining or controlling for the possible interaction between prior trait anxiety and environmental and situational factors during the data collection period (cf. Wood and Gadow, 2010). It was also identified that the PDA format, while simple to program, was not very user-friendly, and this may have had a negative impact on the data collected, and future work in this area should utilise more accessible mobile phone technology.

Implications for intervention

Cognitive behavioural therapy (CBT) techniques depend upon a person being able to identify and monitor their thoughts. This study found that, although not associated with feelings of anxiety, people with AS reported more visual thoughts than NTs, which corresponds with previous studies (e.g. Hurlburt et al., 1994). Clinicians should consider the possible preference towards using visual imagery and the difficulty that this may present some people with AS when trying to translate visual thoughts into words. However, although there seemed to be a tendency to think in a much more visual manner, many of the thoughts recorded by people with AS were described as if they were talking to themselves. This suggests that people with AS are able to engage in internal dialogue, but may require more time or the use of drawing or other visual aids to help translate their thoughts to a verbal description. One participant described their thoughts and feelings in the following way:

There is little point in asking me how I am feeling while I am feeling it. I can take me a day or two to work it out. My brain seems to take sort of videos under certain (generally stressful) circumstances, so that I can work out what was going on later. This makes it hard for me to react appropriately at the time.

In this context, the present findings support the call for what can be termed cognitively informed behavioural therapy (cBT) with people with AS (Hare, 2013)

This study used ESM to collect data in real time and bypass some of the problems with autobiographical memories that have been reported in people with AS. It was apparent that participants could use the technology to record their experiences after a few practice questionnaires and most reported that they enjoyed engaging with the technology with part of the attraction for taking part of in the study being the opportunity to use a palm pilot. On this basis, a follow-on study utilising a modified ESM programme to deliver simple stress management techniques in real time has been developed (Gracey et al., in preparation).

Conclusion

This study found that the prevalence of state anxiety in people with a formal diagnosis of AS was significantly higher than in NT controls and that the cognitive processes associated with state anxiety in people with AS in this study were high levels of self-focus, Type 1 worry, and periods of rumination lasting over 10 min. These processes are components of cognitive models of generalised anxiety disorders, such as the self-regulatory executive function (S-REF) model (Wells and Matthews, 1996) and support using CBT techniques with people with AS. However, the detailed data from this study reveal important considerations specific to people with AS, namely, that most anxious feelings are not associated with anxious thoughts, suggesting that cognitive approaches may first have to help people with AS recognise and articulate thoughts that are linked to anxiety or that more behaviourally weighted approaches may be advantageous in people with AS. In particular, the finding that the AS group reported most of their thoughts as visual imagery suggests that clinicians may need to be creative in this regard.

Footnotes

Appendix

Appendix 1.

Palm Pilot Questionnaire.

Question	Answer type	Area of interest	Coding type
1. What were you thinking about just before the alarm sounded?	Text entry	Content of thought	Coded
2. How would you describe the thought you were having just before the alarm sounded?	Multiple choice confusing annoying worrying practical pleasurable comforting neutral	Appraisal of thought	Numeric/coded (if other is selected)
3. What was the form of the thought you were having?	Multiple choice like written symbols like written text like someone was talking to me like I was talking to myself like an image unsure	Form/description of thought	Numeric/coded (if other is selected)
4. Was this a normal thought for you?	Multiple choice yes no unsure	Whether thought was typical for the person	Numeric
5. How long had you been thinking about this before the alarm sounded	Multiple choice <1 min 1–5 min 5–10 min over 10 min	Amount of time concerned with thought	Numeric
6. How often do you think about this?	Visual analogue 1 = never 99 = constantly	Regularity of thought	Numeric
7. How happy were you feeling just before the alarm sounded?	Visual analogue 1 = very sad 99 = very happy	Mood	Numeric
8. How nervous were you feeling just before the alarm sounded?	Visual analogue 1 = very relaxed 99 = very nervous	Anxiety	Numeric
9. How many people were you with just before the alarm sounded?	Multiple choice I was by myself I was with one other person I was with 2 or more people	Whether anxious thinking is occurring more alone or in company	Numeric
10. How much did the alarm upset you?	Visual analogue 1 = not at all 99 = a great deal	Whether alarm caused distress	Numeric
11. Did the interruption to your thinking upset you?	Multiple choice Yes No	Whether interruption to thinking causes distress	Numeric
12. In what way did the interruption to your thinking upset you	Multiple choice It made me feel nervous I lost my train of thought It annoyed me Other (text entry)	Whether interruption was described as anxiety provoking or other	Numeric/coded (if other is selected)
13. What were you doing just before the alarm sounded	Text entry	Description of behaviour	Coded

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Abell

Hare

(2005) An experimental investigation of the phenomenology of delusional beliefs in people with Asperger’s syndrome. Autism 9: 515–531.

Baron

Groden

. (2006a) Stress and Coping in Autism. New York: Oxford University Press.

Baron

Lipsitt

Goodwin

(2006b) Scientific foundations for research and practice. In: Baron

Groden

. (eds) Stress and Coping in Autism. New York: Oxford University Press, 42–67.

Berthoz

Hill

(2005) Reliability of the Bermond-Vorst Alexithymia Questionnaire. Data from adults with autism spectrum disorder, their relatives and normal controls. European Psychiatry 20: 291–298.

Bowler

Gardiner

Grice

(2000) Episodic memory and remembering in adults with Asperger syndrome. Journal of Autism and Developmental Disorders 30(4): 295–304.

Cardaciotto

Herbert

(2004) Cognitive behavior therapy for social anxiety disorder in the context of Asperger’s syndrome: a single-subject report. Cognitive and Behavioral Practice 11: 75–81.

Cederlund

Hagberg

Billstedt

. (2008) Asperger syndrome and autism: a comparative longitudinal follow-up study more than 5 years after original diagnosis. Journal of Autism and Developmental Disorders 38: 72–85.

Clough

Casey

(2011) Technological adjuncts to enhance current psychotherapy practices: a review. Clinical Psychology Review 31: 279–292.

Department of Health (2010) ‘ Fulfilling and rewarding lives’: The Strategy for Adults with Autism in England. London: Department of Health.

10.

DeVries

Caes

CIM

Delespaul

PAEG

(2001) The Experience Sampling Method in Stress and Anxiety research. In: Griez

EJL

Faravelli

Nutt

Zohar

(eds) Anxiety Disorders: An Introduction to Clinical Management and Research. New York: John Wiley and Sons Ltd.

11.

Dubin

(2009) Asperger Syndrome and Anxiety: A Guide to Successful Stress Management. London: Jessica Kingsley Publishers.

12.

Dunn

Whetton

. (1982) The British Picture Vocabulary Scale. Windsor: NFER-Nelson.

13.

Endler

Kocovski

(2001) State and trait anxiety revisited. Journal of Anxiety Disorders 15(3): 231–245.

14.

Engstrom

Ekstrom

Emilsson

(2003) Psychosocial functioning in a group of Swedish adults with Asperger syndrome or high-functioning autism. Autism 7(1): 99–110.

15.

Fitzgerald

Bellgrove

(2006) The overlap between alexithymia and Asperger’s syndrome. Journal of Autism and Developmental Disorders 36(4): 573–576.

16.

Gaus

(2011) Adult Asperger syndrome and the utility of cognitive behavioral therapy. Journal of Contemporary Psychotherapy 41(1): 47–56.

17.

Gillott

Standen

(2007) Levels of anxiety and sources of stress in adults with autism. Journal of Intellectual Disabilities 11(4): 359–370.

18.

Goddard

Howlin

Dritschel

. (2007) Autobiographical memory and social problem-solving in Asperger syndrome. Journal of Autism and Developmental Disorders 37(2): 291–300.

19.

Gracey

Wood

Hare

(in preparation) Anxiety in Asperger’s syndrome: intervention in real-time.

20.

Grandin

(2006) Stopping the constant stress: a personal account. In: Baron

Groden

. (eds) Stress and Coping in Autism. New York: Oxford University Press, 71–81.

21.

Hair

Black

Babin

. (1995). Multivariate Data Analysis: With Readings. 4th ed. New Jersey: Prentice Hall College Division.

22.

Happé

(2003) Theory mind and the self. Annals of the New York Academy of Science 1001: 134–144.

23.

Hare

(1997) The use of cognitive-behavioural therapy with people with Asperger syndrome: a case study. Autism 1(2): 215–225.

24.

Hare

(2013) Developing psychotherapeutic interventions with people with autism spectrum disorders. In: Taylor

Lindsay

Hastings

. (eds) Psychological Therapies for Adults with Intellectual Disabilities. Chichester: John Wiley & Sons, Ltd, 193–206.

25.

Hare

Mellor

Azmi

(2007) Episodic memory in adults with autistic spectrum disorders: recall for self versus other experienced events. Research in Developmental Disabilities 28: 311–329.

26.

Hektner

Schmidt

Csikszentmihalyi

(2006) Experience Sampling Method: Measuring the Quality of Everyday Life. London: SAGE Publications.

27.

Hintzen

Delespaul

PAEG

van Os

. (2010) Social needs in daily life in adults with Pervasive Developmental Disorders. Psychiatry Research 179(1): 75–80.

28.

Hofvander

Delorme

Chaste

. (2009) Psychiatric and psychosocial problems in adults with normal-intelligence autism spectrum disorders. BMC Psychiatry 9: 35–44.

29.

Hurlburt

Happeé

Frith

(1994) Sampling the form of inner experience in three adults with Asperger syndrome. Psychological Medicine 24: 385–395.

30.

Jennes-Coussens

Magill-Evans

Koning

(2006) The quality of life of young men with Asperger syndrome. Autism 10(4): 403–414.

31.

Kim

Szatmari

Bryson

. (2000) The prevalence of anxiety and mood problems among children with autism and Asperger syndrome. Autism 4(2): 117–132.

32.

Lawrence

Alleckson

Bjorklund

(2010) Beyond the roadblocks: transitioning to adulthood with Asperger’s disorder. Archives of Psychiatric Nursing 24(4): 227–238.

33.

Moree

Davis

(2010) Cognitive-behavioral therapy for anxiety in children diagnosed with autism spectrum disorders: modification trends. Research in Autism Spectrum Disorders 4(3): 346–354.

34.

Myin-Germeys

Birchwood

Kwapil

(2011) From environment to therapy in psychosis: a real-world momentary assessment approach. Schizophrenia Bulletin 37(2): 244–247.

35.

Myles

Ferguson

Hagiwara

(2007) Using a personal digital assistant to improve the recording of homework assignments by an adolescent with Asperger syndrome. Focus on Autism and Other Developmental Disabilities 22(2): 96–99.

36.

Palmier-Claus

Myin-Germeys

Barkus

. (2011) Experience sampling research in individuals with mental illness: reflections and guidance. Acta Psychiatrica Scandinavica 123(1): 12–20.

37.

Reis

Gable

(2000) Event-sampling and other methods for studying everyday experience. In: Reis

Judd

(eds) Handbook of research methods in social and personality psychology. New York: Cambridge University Press, 190–222.

38.

Renty

Roeyers

(2006) Quality of life in high-functioning adults with autism spectrum disorder. Autism 10: 511–524.

39.

Skokauskas

Gallagher

(2010) Psychosis, affective disorders and anxiety in autistic spectrum disorder: prevalence and nosological considerations. Psychopathology 43(1): 8–16.

40.

Sterling

Dawson

Estes

. (2008) Characteristics associated with presence of depressive symptoms in adults with autism spectrum disorder. Journal of Autism and Developmental Disorders 38: 1011–1018.

41.

Sze

Wood

(2007) Cognitive behavioral treatment of comorbid anxiety disorders and social difficulties in children with high-functioning autism. Journal of Contemporary Psychotherapy 37: 133–143.

42.

Van Steensel

FJA

Bögels

Perrin

(2011) Anxiety disorders in children and adolescents with autism spectrum disorders: a meta-analysis. Clinical Child and Family Psychology Review 14: 302–317.

43.

Wechsler

(1999) Wechsler Abbreviated Scale of Intelligence. San Antonio, TX: Psychological Corporation.

44.

Wells

(1997) Cognitive Therapy of Anxiety Disorders: A Practice Manual and Conceptual Guide. Chichester: Wiley.

45.

Wells

Matthews

(1996) Modelling cognition in emotional disorder: the S-REF model. Behaviour Research and Therapy 34(11): 881–888.

46.

Weiss

Lunsky

(2010) Group cognitive behaviour therapy for adults with Asperger syndrome and anxiety or mood disorder: a case series. Clinical Psychology & Psychotherapy 17: 438–446.

47.

White

Oswald

Ollendick

. (2009) Anxiety in children and adolescents with autism spectrum disorders. Clinical Psychology Review 29: 216–229.

48.

Wood

Gadow

(2010) Exploring the nature and function of anxiety in youth with autistic spectrum disorder. Clinical Psychology: Science and Practice 17(4): 281–292.

49.

Zigmond

Snaith

(1983) The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica 67: 361–370.