Decisiontry Autism and Decision Making
Autism and Decision Making
Interview with Dr Lydia Luke by Lorna Rouse
Dr Lydia Luke
Lydia is a Clinical Psychologist at Oxfordshire County Council and completed her PhD at the Cambridge Intellectual and Developmental Disabilities Research Group.
So if I can start by asking what made you interested in the area of autism and decision-making?
Well, I did my undergraduate degree in Experimental Psychology here at Cambridge and during that time, I was lectured by Simon Baron-Cohen, who is an expert on autism. That was the first time I was introduced to and became interested in autism as a condition. When I finished university, I worked as a Teaching Assistant with children with Asperger's and this experience increased my interest in the condition. I'd also been interested in decision-making and other cognitive processes as an undergraduate. However, the first project I worked on as a Research Assistant was about the Mental Capacity Act and it was this that really inspired me to think about decision-making in a way that had very practical applications. My supervisors (Dr Isabel Clare and Dr Howard Ring) were interested in real-life decision-making and had a lot of expertise in autism, so these seemed like good topics to bring together. In trying to review the literature, I realised that very little is known about how people with autism make decisions or whether they have particular difficulties. This contrasts with learning disabilities, dementia and schizophrenia, for which there's been quite a lot of research.
What does the mental capacity act say about autism and decision-making?
Well it doesn't, and shouldn't, say anything very specific about autism. However, it does say that everybody is presumed to have the capacity to make decisions for themselves unless they have an impairment or disturbance of the mind or brain that can affect their ability to understand, weigh up, retain or communicate the relevant information. So although there's nothing specific about autism, it raises the question of whether, for some people who have autism, their condition means that they might have an impairment in one or more of these areas.
How did you think decision-making might be different for people with autism?
Well, to start with, there are a few autobiographical accounts of people with autism, which suggest that decision-making can be difficult and overwhelming. Secondly, there are clinical case studies of how some people with autism, with strong special interests, appear to make bad decisions in order to pursue these. For example, there is a case of a man with autism who got into trouble with the law because he was trespassing to look at washing machines, which were his special interest. Thirdly, autism is known to be associated with some cognitive differences, such as difficulty planning and focusing attention, and these differences have been shown in other clinical groups to impair decision-making. There are also some differences in brain regions that appear to underpin emotional involvement in decision-making. Finally, we knew that people with autism are often very anxious, and that high levels of anxiety can make decision-making more difficult. To identify whether there are specific ways that decision-making might be affected by autism, we carried out an exploratory survey of people with autism, and family members and support workers of people with autism. 120 people with autism took part, and the findings indicated that the participants were often exhausted by decision-making, found it stressful, took a long time to make decisions, and had a tendency to spend time gathering information rather than committing to a decision.
What were the main questions you were trying to answer when you started your PhD?
To start with it was rather broad - how might decision-making be affected by autism and is there a way to assess whether someone's special interest might be affecting their ability to make decisions? However, addressing these questions turned out to be very challenging and did not assess the main experiences identified in my survey. My research questions were then centred upon 1) whether the experiences reported in the survey were specific to people with autism; 2) whether there was any empirical evidence for these experiences; and 3) whether there were differences in the decision-making processes of people with autism that could account for their experiences. Ultimately I wanted my research to lead to a better understanding of how decision-making might be affected by autism, and to find out if there are ways that people with autism could be better supported.
You used a combination of qualitative and experimental methodology. Could you briefly outline both those methods?
My survey was intended mainly to be quantitative data, so I used a questionnaire that went out online and by post to members of the National Autistic Society and other autism-support groups. I asked people how difficult they might find particular types of decisions, how often they experienced particular problems in decision-making, whether they thought their autism gave them strengths when it came to decision-making, and how frequently they thought they benefited from the condition. The questionnaire also included some open text boxes at the end of these lists for people to give examples and provide more information. The written responses were surprisingly detailed, and some people had some very insightful things to say about their difficulties, the situations in which they occurred, as well as their strengths. In particular, several participants said how they thought that their autism made it easier for them to use logic when it came to decision-making, and that they had a tendency to see things in black and white, which could made decision-making easier.
So autism had a positive effect as well?
Yes it did, although the responses were mostly negative about the effects of autism on decision-making. I used a thematic analysis approach for my questionnaire data because it was an exploratory study and intended to be the starting point for the second phase of my study. In the second phase, I used some questionnaires and established tasks of decision-making to address specific questions that substantiated or explored the survey findings. So, for example, I looked at whether there were differences in the everyday decision-making experiences of people with autism, compared to the general population, and whether people with autism really did take longer to make decisions. I also looked at whether people with autism had a tendency to gather more information prior to making decisions. The second phase also included a very well known task, the Iowa gambling task, because a previous study had found a few interesting trends using this task with a small sample of people with autism. My study provided a good opportunity to test these out in a much larger sample.
What had the previous study found?
They found that people with autism responded rather erratically on the task where you have to choose from four decks of cards in order to win points. They also found a trend towards people with autism being more influenced when they lost money rather than when they won money, compared to the general population.
So does the general population usually respond more to winning?
Not necessarily, but when participants with autism and the general population were compared, it looked like the people with autism were more sensitive to negative outcomes. We wondered if this could account for their generally negative perceptions of decision-making; if the bad experiences were more salient than the times when decisions had worked out well. However, we didn't find any evidence that the participants with autism were more sensitive to loss than the general population.
Who were your participants and were there any particular challenges working with people with ASDs?
They were intellectually able adults who had been diagnosed with an autism spectrum condition and they were recruited through volunteer databases and local and national support organisations for people with autism. The controls were adults, matched for IQ, who were recruited from the local general population. I think recruitment is a tricky area because, whilst some people with autism are very keen to take part in research, there may be lot of people with autism who, because of their social impairments, are not and these people might not be very forthcoming in responding to adverts. So, you do worry whether you are getting a very special group of people who are happy to help with research. I don't think I had any particular difficulties in working with the participants with autism. I drew on my experience of working with people with autism and made an effort to create a quiet and relaxed environment that did not put pressure on the participants. Recruiting controls was actually more of a challenge!
Okay, so could you tell me about your main findings?
The main finding from the survey was that people with autism perceive difficulties in everyday decision-making, although they also perceive some strengths associated with the condition, such as a good eye for detail and an ability to think things through logically. My empirical findings support some of the experiences that they were reporting. We found evidence that people with autism took longer to make decisions in the laboratory, and that they tended to sample information until they were more likely to make a correct decision than the general population. Another finding was that, on two of my tasks, the people with autism made more logical choices, supporting what they'd said about autism perhaps giving them an advantage when it comes to using logic. I have to stress, however, that my research treated participants with autism as a single group and there's a lot of diversity amongst people with the condition.
Do you think there might be any differences in decision-making within that spectrum?
Quite possibly, I think my research is very much the starting point for looking at decision-making in autism and there's a lot of work to be done looking at individual differences within the group. For example, some people spent much longer on their decisions than others and we don't know if there are relationships with IQ or with the severity of the condition that might explain that.
If you could look further at anything in this area, is there anything that you'd particularly like to focus on?
If I could run more experiments, I'd be interested in the time course of decision-making in autism. In my study, people with autism were given as much time as they wanted to make the decisions in, and, generally, the groups made similar decisions, but the participants with autism took longer. It would be interesting to have different conditions where the amount of time people have to make that decision is restricted. That could give an indication of whether taking longer is just a style of decision-making, or whether decision-making processes are happening more slowly. If you restrict the time given to people, does the quality of the decisions remain the same? Or is this something that people with autism would find particularly stressful?
So there are lots of questions still to be answered around autism and decision-making?
Yes. One of the limitations of my study was that decision-making was looked at in a very restricted context; participants were in a laboratory, completing tasks presented on a computer, and making decisions that are quite abstract. For example, the laboratory decisions often involve weighing up points and probabilities of winning and losing. Although they're very clever tasks that have been designed to capture elements of real life decision-making, they do lack ecological validity and it would be interesting to know more about how people with autism make decisions in a more naturalistic environment. If you could do that, you may perhaps see more of the experiences people report being substantiated by experimental work. We've established that there are some differences in the way that people with autism make decisions compared to the general population, but our work doesn't really address the mechanisms underpinning those differences. Understanding those mechanisms is the next step.
You mentioned that you were interested in the practical applications of decision-making. What are the practical implications of your findings for people with autism and those who support them?
The practical implications are that people with autism could perhaps benefit from being given more time to make decisions, and reassurance that they are able to make these decisions. We certainly didn't find any evidence of poor quality decision-making. Those who support people with autism may also be able to help by tailoring the decision-making process to the person's strengths. For example, in education decisions, where there is a lot of available information about the different options, it may be helpful to gather the relevant information for the person with autism so that they don't become bogged down in information gathering and can focus on the analytical part of the process that they're likely to be quite good at. I think the ultimate take home message from my work is that it shouldn't be assumed automatically that able people with autism might not benefit from some support when it comes to making decisions. It certainly seems that decision-making is not an activity that is very enjoyable for people with autism and it can cause a great deal of stress.
What are the kinds of everyday situations that are particularly difficult?
Probably the prime example of where people were having difficulty was shopping; being presented with different choices for very similar or the same items and having to choose one over the other seemed to be a massive cognitive task. The participants reported that they were good at identifying all the different variables that could be considered, but it just became a long list and then, of course, quite a difficult decision. Thinking about the ultimate message to carers, I think it's also important to realise that everyone with autism is an individual and not everyone is going to need support. Some people will be very capable, so efforts to provide support really do need an initial assessment about the person's likely strengths and weaknesses. My research just highlights the fact that it shouldn't be assumed that able people with autism will have no difficulties. Support may still be beneficial.
That's great, thank you very much.
This interview originally appeared in News & Views June 2011.