Exploring the autistic and police perspectives of the custody process through a participative walkthrough

Background: Research suggests that autistic individuals may be more likely to come into contact with police and have more negative experiences in police custody. However, limited information about the difficulties they experience during the custody process is available. Aims: This study explores the experiences of autistic individuals and officers during a walkthrough of the custody process to identify specific difficulties in these encounters and what support is needed to overcome these. Methods and procedures: A participative walkthrough method was developed to provide autistic individuals and officers an interactive opportunity to identify areas where further support in the custody process was needed. Two autistic participants and three officers took part in the study. Outcomes and results: Autistic participants reported negative experiences due to: i) the emotional impact of the physical setting and custody process ii) communication barriers leading to increased anxiety and iii) exposure to sensory demands. Officers highlighted three factors which limit their ability to support autistic individuals effectively: i) the custody context ii) barriers to communication and iii) knowledge and understanding of autism. Conclusions and implications: Adjustments are needed to the custody process and environment to support interactions between autistic individuals and officers and improve the overall wellbeing of autistic individuals. What this paper adds? There has been little in-depth research into the lived experiences of autistic individuals throughout the custody process. Previous research has not explored the difficulties which autistic individuals might experience at specific points of the custody process or what adjustments might support the interactions between autistic individuals and officers to improve overall wellbeing. This study is one of the first to examine the experiences of autistic individuals during the custody process. A novel walkthrough technique using a combination of qualitative methods was employed to gain an insight into police and autistic interactions and experiences of the custody process. It builds upon the evidence that autistic individuals have negative experiences in police custody by identifying specific parts of the custody process that create difficulties for autistic detainees. The research also further explores the factors which may affect the ability of officers to support autistic detainees effectively. https://doi.org/10.1016/j.ridd.2019.103545 Received 14 June 2019; Received in revised form 11 November 2019; Accepted 15 November 2019 ⁎ Corresponding author. E-mail address: chloe.holloway1@nottingham.ac.uk (C.A. Holloway). Research in Developmental Disabilities 97 (2020) 103545 0891-4222/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/). T


Introduction
Research suggests that autistic individuals are more likely to come into contact with police than the general population (Debbaudt & Rothman, 2001). For example, Tint, Palucka, Bradley, Weiss, & Lunsky (2017) reported that 16 % of 284 autistic participants had experienced police contact over a 12-18 month period (Tint et al., 2017). Some studies have suggested that this could be due to autistic individuals being more likely to commit a criminal offence or engage in certain types of criminal behaviour (see Allely & Creaby-Attwood, 2016 for discussion). However, other studies have disputed this, reporting that autistic individuals are in fact no more likely than the general population to commit an offence (see King & Murphy, 2014 for review). Alternative explanations for the increased risk of autistic individuals coming into contact with police have been reported by other studies. For instance, it has been suggested that autistic individuals are at greater risk of being victims of a crime (Brown-Lavoie, Viecili, & Weiss, 2014). In addition, another study has also reported that autistic individuals may be at greater risk of their behaviour being misinterpreted by police, resulting in an increased likelihood that they may be arrested for a criminal offence (Dickie, Reveley, & Dorrity, 2018).
Police custody plays an integral role in the criminal justice system in responding to and dealing with individuals who have been arrested in connection with offending behaviour. In England and Wales, when an individual has been arrested by a police officer on reasonable suspicion that they have committed, are committing, or about to commit a criminal offence, they may be brought to police custody (see s 24 Police and Criminal Evidence Act 1984 ('PACE')) until a decision is made about whether they are to be charged with the offence. A suspect may be detained in police custody for up to 24 hours without charge for the purposes of investigating the offence (see s 37(1) and s 41 PACE). Many decisions made in the custody context can influence the personal and legal outcomes of individuals. For instance, the detainee can decide to exercise the right to legal representation or the right to silence, or they may choose to admit guilt and/or accept a police caution. The custody officer oversees many of these decisions as they are responsible for ensuring detainees are treated according to the main legislative safeguards (see s 39(1)(a) PACE). In particular, the custody officer is responsible for informing detainees of their rights and ensuring the welfare of individuals whilst in custody.
To some extent, all individuals will experience difficulties in police custody as suspects will be subject to a series of intrusive processes including booking-in, fingerprinting, DNA swabbing, drugs testing and police interview (see Skinns, 2011 for discussion). They may also be detained in a cell for a significant period of time while the investigation takes place and they may experience a loss of privacy, isolation and a loss of control (see Skinns, 2011 for discussion). This can sometimes result in adverse outcomes (see Skinns, 2011 for discussion). However, the addition of sensory and communication differences in autism may exacerbate the impact of these factors further, making them more vulnerable to adverse outcomes (Chown, 2010;Crane, Maras, Hawken, Mulcahy, & Memon, 2016;Woodbury-Smith & Dien, 2014). Specifically, if important legal information is not conveyed in an accessible way, this could lead to autistic detainees making ill-informed decisions in custody. In addition, barriers to communication between neurotypical and autistic individuals may prevent adjustments to the custody process or environment from being made which could further impact negatively on their welfare. Despite the increased likelihood of coming into contact with the criminal justice system, police officers and other criminal justice professionals report feeling ill-equipped to adequately support autistic individuals (Crane et al., 2016;Dickie et al., 2018). But few studies have explored autistic individuals' experiences in the system. Working in partnership with autistic individuals to gain their views on how to enhance service provision (Robertson, 2010), and to ensure research is both meaningful and valuable to the autistic community (Fletcher-Watson et al., 2019) can lead to better translation of the research findings to real-world settings, improving outcomes for autistic individuals.
Of the few studies that have explored the criminal justice system from an autistic perspective, all report that autistic individuals have negative experiences in police custody and are generally dissatisfied with their interactions with police (Allen et al., 2008;Crane et al., 2016;Helverschou, Steindal, Nøttestad, & Howlin, 2018). As part of a larger study into the prevalence of offending behaviour among autistic individuals in South Wales, Allen et al. (2008) conducted interviews with six autistic individuals who had experience of the criminal justice system. They documented several difficulties including "not being able to take everything in; feeling in the spotlight; not knowing what was going to happen next, and feeling uncomfortable with the other people at the police station ..." (Allen et al., 2008: 754). Following an online survey of both police officers (n = 394), autistic individuals (n = 31) and their families (n = 49) in England and Wales, Crane et al. (2016) also reported that autistic individuals experienced emotional stress and breakdowns in communication in custody. In part, this was attributable to the inappropriateness of the physical setting of the interview room and custody suite and a lack of appropriate support. This study also found that some autistic individuals may not disclose their diagnosis due to a fear of being victimized or discriminated against. Helverschou et al. (2018) interviewed nine autistic individuals who had been convicted of a criminal offence. Although the majority of the research focused on reasons underlying the offending behaviour and experience in prison in Norway, some questions touched on their experiences during the custody process. While the processes in Norway may be different to England and Wales, it is interesting that the individuals reported similar negative experiences to those reported by participants in the UK. Specifically, they reported being confused about the reasons for their arrest or concerns about the lack of a legal representative that understood them.
While these studies provide evidence for the need of greater support for autistic individuals within the criminal justice system, there are some limitations to the methodological approaches used. Interviews and surveys conducted after a period of time has passed may fail to capture important details about specific parts of police procedures that are significant barriers to participation in the criminal justice system. The sample of participants who complete online surveys may also not fully reflect the broader range of experiences as they may be disproportionately completed by those who are most dissatisfied by their experiences. Furthermore, these studies explored autistic experiences of the criminal justice system more broadly. They did not ask participants to reflect on specific aspects of the custody process or identify what support may be required to allow them to participate effectively. To address this, the current research will investigate the custody process to gain detailed information about potential changes that could make the process more accessible for autistic individuals. Crane et al. (2016) suggested it would be valuable to carry out direct observations of how police procedures are conducted with autistic individuals. This would allow the researcher to note specific areas of difficulty and when they occur, which might be missed in subsequent interviews with participants. This approach has further value in that it would allow researchers to obtain perspectives from both autistic individuals and the officers interacting with them. This allows us to address the double empathy problem (Milton, 2012;Milton, Heasman, & Sheppard, 2018;Sheppard, Pillai, Wong, Ropar, & Mitchell, 2016) which argues that breakdowns in communication are a two-way process. Specifically, just as neurotypical individuals may struggle to understand the intentions of nonautistic people, difficulties may also stem from an inability of neurotypical individuals to infer what autistic individuals are thinking (Sheppard et al., 2016).
Therefore, it was the aim of this research to carry out a participative walkthrough which used a combination of direct observation and interview techniques to identify barriers that might affect an autistic individual's participation in the custody process. Two autistic individuals and three officers took part in the study. The objectives were to: i) identify specific points in the custody process where adaptations are needed, ii) assess how autism impacts upon the custody process, and iii) identify recommendations for changes to the custody process/environment that would improve the experiences of autistic individuals.

Participative walkthrough
Due to the practical, ethical and legal limitations associated with carrying out an observation of autistic individuals detained in police custody (see Demarée, Verwee, & Enhus, 2013), it was considered more appropriate to simulate this process using a walkthrough method. The participative walkthrough is an experimental method adopted for the purposes of this research. It builds on the basic principles of the 'go-along' method which is typically used by researchers to learn more about a neighbourhood or place, or experimentally, to explore new and unfamiliar situations (Carpiano, 2009;Kusenbach, 2003). This method combines two different qualitative methods: field observation and interviewing (Carpiano, 2009). During observations, people do not usually provide a verbal commentary on their activity, making it difficult to access their concurrent experiences and interpretations (Kusenbach, 2003). Similarly, interviews keep informants from engaging in their usual activities in the environments where they naturally occur meaning important aspects of that lived experience may be overlooked (Kusenbach, 2003). In contrast, by using a go-along approach researchers can observe an individual's responses in context and simultaneously access their experiences at a specific time or location (Kusenbach, 2003). Researchers are therefore able to explore the informant's experiences, interpretations, and practices using a combined approach of asking questions and observation (Carpiano, 2009).
The participative walkthrough goes beyond the traditional 'go-along' to recreate an interactive process. This study was designed to provide officers and autistic individuals the opportunity to actively engage in the custody process and with each other, rather than simply being passive observers to what would take place in custody. Another layer was added by the researchers who observed the walkthroughs assessing the participants' behavioural and emotional responses. These observations also helped guide the subsequent interviews and provide additional context to themes identified in the data, allowing the researchers to identify points to discuss in the interview and verify the accuracy of what was reported. The researcher also served a supportive role for the walkthrough and was there to guide the process and intervene, if needed.

Participants
Ethical approval for the study was granted by the ethics committee at the School of Law, University of Nottingham. Autistic participants were recruited on the basis they: i) were resident in the UK, ii) were over the age of 18, iii) had a diagnosis of an Autism Spectrum Condition (including Asperger's Syndrome), iv) did not have previous experience of being detained in police custody, and v) had sufficient communication skills to take part in the research. Two autistic individuals, one female aged 24 (referred to as Anne) and one male aged 33 (referred to as Ben), took part. Both were British nationals with English as their first language. One participant lived independently and both had educational attainment through to University level. Nottinghamshire police were the focus of the study because of their proximity to the University and interest in developing their vulnerability strategy further. One custody sergeant and two detention officers were recruited via internal email sent by the custody and development officer to assist with the walkthrough. The officers were not required to have any training or experience of autism, although one reported having personal experience of interacting with autistic individuals outside custody. Informed consent was obtained from all participants prior to the study.

Procedure
The walkthrough focused on three parts of the custody process: i) booking-in ii) processing and iii) cell detention. The walkthrough was carried out in an operating police station on an overflow floor which was not currently in use. Two walkthroughs were conducted. The walkthroughs lasted 51 and 53 min, respectively. These were audio recorded with the consent of participants. Two officers, the autistic participant and CH were present during each walkthrough. DR was also present during one walkthrough to provide additional support to the autistic participant. CH took notes on the quality of interactions between individuals (i.e. ease of communication), and recorded any visual emotional and behavioural cues (e.g. self-stimulatory behaviour) showed by the participants in response to what was happening, and provided support to both participants throughout the process. Prior to the walkthrough, participants read an information sheet and consent form which they signed and returned to the researcher. They were also given an overview of what would happen during the study and informed that each process was optional. Consent was verbally confirmed at each stage. Participants were allocated a scenario to contextualise the study (theft or criminal damage) which was explained to them.

Booking-in
Each autistic participant was asked to answer a series of background questions set out by standard police procedures. The officers answered any queries raised by the participants and explained the purpose of the booking-in process. Participants were then asked to undergo a personal search. The officers explained what would happen and showed them the items used in this process. Where the participants consented, the search was then carried out by an officer of the same gender.

Processing
The autistic participant was then taken to the processing room located in an operating part of the station. No other detainees were present at this time. The different processing procedures were explained accordingly: i) UV Smartwater test, ii) fingerprinting, iii) DNA test, iv) drug test and v) photographs. The custody sergeant also explained what safeguards would be in place and any legal implications. The officers simulated the Smartwater test, fingerprinting and photographs where the participant consented.

Cell detention
Finally, the autistic participant was taken to look around a holding cell, interview room and shower room. They were specifically asked to talk about how they felt about the cell. The officers demonstrated how the intercom in the cell operated, what it would be like if the door was closed and what would happen during a routine check.

Interviews
The participant interviews were semi-structured and conducted following an interview schedule to allow for consistency between individuals. All interviews were audio recorded and notes were taken throughout. DR conducted interviews with each autistic participant. One interview lasted 47 min and the other was 64 min long. Questions were divided into sections: i) overall study ii) booking-in iii) processing iv) cell detention and v) final comments. All questions were open-ended asking the participants to describe their experiences of each process and identify any difficulties with or ways of improving these processes (see Appendix A). CH also carried out a joint interview with the custody sergeant and one detention officer. The other detention officer was unavailable for interview after the study. The interview was 25 min long. General open-ended questions were asked about the officers' experiences of taking part in the study and interacting with the autistic participants (see Appendix B).
Upon completion, all participants were de-briefed and invited to discuss any concerns. Researchers maintained contact with the participants for a limited time after the study to check if they had further questions and ensure they were not experiencing any negative consequences from their participation.

Data analysis
The walkthrough and interviews were transcribed in full and analysed by CH using a combination of inductive and deductive approaches to thematic analysis (Braun & Clarke, 2006). This allowed the researchers to identify themes relating specifically to the officers or autistic participants and better compare their perspectives. Initially an inductive approach was taken due to the novelty of the walkthrough method and the exploratory nature of the research questions, which allowed us to code directly from the data to reflect the actual experiences of the participants. Each transcript was read line by line and colour coded according to identify specific themes. These initial codes were then reviewed, adapted and categorised into overall themes to better describe what was happening in the data. The second stage of analysis involved a more deductive approach as the transcripts and field notes were considered alongside each other to compare the experiences of the participants and allow the researcher to draw inferences from the data that were not explicitly articulated. Specifically, we drew upon the theoretical construct of "double empathy" (Milton, 2012;Milton et al., 2018) to help us better frame relationships between the data. To verify the accuracy of the final analysis, the data was subsequently reviewed and interpreted independently by NM and DR. A few discrepancies were resolved prior to the identification of final themes.

Autistic perspective
Three core themes were identified: i) impact of the custody environment ii) the contribution of communication difficulties to anxiety and iii) impact of sensory sensitivities on autistic individuals' experiences in custody (see Table 1 for summary).

Impact of the custody environment
Both participants were concerned about what it would be like to be detained longer in police custody. This was linked with the physical surroundings and the emotional impact of experiencing the custody process.
3.2.1. The effect of the physical surroundings on well-being 3.2.1.1. Unfamiliarity. The unfamiliarity of the custody environment was a concern: "(…) it's mainly the novelty of the environment -the unfamiliarity of it" (Ben interview). This was partially attributable to the fact it was the participants first time in police custody.

Small spaces.
The lack of space was also identified as a potential difficulty: "very claustrophobic" (Anne interview). Ben shared concerns about the small space particularly over a longer period of time with nothing to do: "I think if I was in there for a whole day… I think I'd be very claustrophobic… (…) I wouldn't have been doing much constructive activity. (…) I'm worried about perhaps the space to move (…) perhaps, the exercising (.." (Ben interview). These concerns extended to the interview room: "a bit… boxy, a bit dingy in places, a bit (…) claustrophobic" (Ben walkthrough). Conversely, the processing room was viewed more positively: "(…) was quite spacious and I'm thinking that others might be tighter, so that helped in there (.." (Ben interview).

Sleep.
The custody environment was also connected to the potential inability to sleep. This was due to the lighting in the cell which is always turned-on: "This doesn't get any darker at night. This is what it'd be like at 2 a.m." (Ben walkthrough). It was suggested this would cause difficulties sleeping which, combined with having nothing to do, would heighten the impact of detention: "if I'd been in here for… what, let's say, a few hours… I think I'd probably start getting bored (…)" (Ben walkthrough).

Privacy.
Both participants also discussed potential difficulties created by the lack of privacy in custody. The open toilet was a key issue: "(…) the toilets I think, I find most intimidating!" (Ben walkthrough). Both participants stressed they would not have used the toilet unless they were "desperate" (Ben interview). This was because they thought they would be visible on the CCTV camera. But this is not always the case: "(…) it's not constantly being watched -but it's constantly being filmed in case something happens and the toilet areas greyed out, so you can't see people on the toilet." (Anne walkthrough).

Understanding of custody staff.
It was stressed that the manner and response of custody staff can help to minimise the emotional impact of custody: "(…) I felt a lot easier about it than I initially thought… and, the officers who we, dealt with today, were… very… understanding, (…) overall, I didn't feel overly intimidated, but (…) specific points (…) could have been a little bit… intimidating." (Ben interview). Importantly, it was suggested knowledge of autism was crucial: "(…) although they… seemed to… uh, try to be understanding they actually… appeared to have not that much knowledge. And so… probably being under more stress… if you were (…) in custody for real. (…) it (…) might be even harder to… feel comfortable to tell them, like, what was wrong. And especially some of the questions, which were key" (Anne interview).

The contribution of communication difficulties to anxiety
Both participants reported feeling less anxious when the officers explained what was happening as it helped them understand the custody process. In contrast, they reported feeling anxious when this was not fully explained, or the language was not clear.

The need for more information about police procedures
Not knowing what was going to happen was a key source of anxiety. This was illustrated during the walkthrough when the Smartwater test was conducted by switching off the normal lighting and turning on the ultraviolet light. It was observed by the researcher that the "sudden change startled [Anne]" (Field notes). This was because the detention officer did not explain what they were doing: "(…) I didn't know what -was gonna happen. (…) all they said was 'Stick your hands out' (…) I didn't know if they were gonna suddenly search me or what." (Anne interview). It was mentioned that in some situations this might cause some autistic individuals to "lash out at someone" (Anne interview).
Anne emphasised that it helped when the officers provided a context for what was happening by explaining what they were going to do and why they had to do each thing. Ben concurred that this helped reduce his anxiety: "(…) it just removed some of the fear about it. You know, when you know what's actually being done… you don't have the anxiety about something that's… impending that isn't." (Ben interview). Anne also felt it might help to be kept informed about what is happening later on in the process and receive regularly updates of any changes if they were detained for longer periods of time. Ben agreed that additional information about what might happen would be useful: "

I think the foresight of what would happen next and possibly when it's likely to happen… (…), step-by-step what's likely to happen and (…) if you're charged, what happens if you're not charged and um… you know, … perhaps explaining some of the (professionals) roles -you know have some literature, because reading material's always good." (Ben interview).
It was also suggested that more accessible forms of information should be provided. For example, "visual prompts of what was gonna happen" or "a visual timetable" (Anne interview) and a booklet outlining the custody process.

The need for clear unambiguous communication 3.3.2.1. Ambiguous questions.
The autistic participants often found it difficult to understand some of the questions asked by the police officers. Open and ambiguous questions played an integral part of this. For example, it was noted by the observer that when Anne was asked "Where do you live?" in the booking-in stage, she only gave the name of the city where she lived. When further prompted by the officer with "Where in (city name)?", Anne only gave the specific area rather than the address. Anne commented in interview "(…)

they asked me (…) where I lived. (…) they could've easily just said 'What's your address?'" (Anne interview).
There was further evidence that the phrasing of other questions during the booking-in led to a failure to disclose important medical information or clinical diagnoses. For instance, when Ben was asked "do you have any injury or illness currently?" he did not disclose a medical condition due to the wording of the question. This was only picked up several questions later when he was asked about whether he was taking any current medication (Field notes). Both participants also reported that they found it difficult to know when it was appropriate to disclose they were autistic: "I wouldn't have mentioned autism… (…) Had they not already known" (Anne interview). This was because there was no direct question on autism, only questions asking "do you suffer from any mental health problems or depression?" and "do you have any learning disabilities that we need to provide any help or support for?".
The phrasing of questions was also mentioned in relation to asking for food and drink. It was explained that standard protocol is to regularly check on the detainee in the cell to check they are okay. Anne stressed that: (…) 'Are you okay?' could be answered in general … I just say yes." (Anne interview). She further commented in interview: "I think, being asked specifically if you want a drink or food [would help] … 'cause it's something I find difficult to ask for (…) in a strange situation" (Anne interview).
It was suggested communication between the autistic individuals and officers could be better supported by making "open questions effectively into closed questions" (Anne interview). Prompt sheets would also have helped: "could have had like, prompt sheets (…) so when they say 'How does autism affect you?', be able to say (…) things that are quite common"; "like lighting, sound. "(Anne interview). This would make it easier for autistic individuals to communicate what specific support they might need and minimise the risk that they may not disclose this information.

Legal jargon.
Technical language (i.e. solicitor) caused additional confusion. This was a specific concern for Anne: "added some anxiety because… (…) I had to re-ask… what it was"; "they didn't seem to understand (…) why I didn't understand" (Anne interview). It was observed that this encouraged her to waive her legal rights during the study (Field notes).

Communicating in the cell.
There were also concerns about using the intercom to contact the officers: "I wouldn't feel comfortable using it. 'cause it's like a phone… I don't like phones...";"'cause you can't see who you're talking to. (…) it's like the anxiety of waiting for someone to pick up"; "there's no body language, (…) you have to rely on… words..","you can't sign or anything" (Anne interview).

The impact of sensory sensitivities on autistic individuals' experiences in custody
Several aspects of the custody environment were identified as creating sensory difficulties. C.A. Holloway, et al. Research in Developmental Disabilities 97 (2020) 103545 3.4.1. Noise As the study was conducted on a quieter floor, there was minimal background noise. However, Ben commented on how he would have felt if the walkthrough had been conducted in busier surroundings: "(…) there wasn't a lot of noise there, but if there was… if there was lots clattering, lots of loud noises, it could have been quite difficult." (Ben interview). Anne also reported hearing "buzzing" coming from the lights and described the cell as "quite echoey" (Anne interview). She emphasised these noises would have been difficult to cope with: "you're already anxious going in. And I would say all of those, sort of, aspects are heightened… and so it's harder to deal with them"; "you can't block them out as easily." (Anne interview). To minimise exposure to noise, it was suggested that it would be better to use a quieter part of the station, if possible (Ben interview). Alternatively, it was thought that people might benefit from "ear-muffs" or ear defenders (Ben interview).

Visual
A potential source of visual sensory stress was the artificial lighting: ""it could throw a few (…) sensory issues after prolonged exposure" (Ben interview). Being able to dim the lighting would have minimised the impact of this. It was also suggested that the lighting should be replaced with "softer lights" or "different colours of lighting" (Ben interview). Colour was also a concern: "a lot of people with autism can be colour sensitive as well -so certain colours." (Ben interview). Duller colours were preferred: "(…) you wouldn't want a cell to be red for instance… because that could, you know, make people… quite angry? (…)" (Ben interview) In addition, it was felt that "softer floor colours" and "more matt" paint would be less demanding (Ben interview).

Tactile
Lastly, Anne and Ben referred to the tactile demands of the custody process. During the study, Ben took part in the search, fingerprinting and photographs. While the search did not create problems at the time, in different circumstances this could have raised sensory issues: "(…) if it wasn't in a such a consensual nature it would have felt a little bit intrusive." (Ben interview). In contrast, Anne opted for a description of this process because she was concerned about the sensory demands involved.
Tactile demands were perhaps most evident during the processing stage for both participants who raised concerns about the invasiveness of the procedure. For example, they referred to the forcefulness of the fingerprinting: "(…) in a busy situation I'd imagine it might have been slightly more forceful, which might have been more irritating. (…)" (Ben interview). This was amplified by the fact that "they said that they could do it against your will (…) Feel sort of forced to do it." (Anne interview). The DNA swab would raise similar issues: "Very uncomfortable. Having stuff in my mouth that I don't want." (Anne interview). It was noted that a further concern was having to wear different fabrics for the photographs, as this may be a "potential sensory trigger" (Field notes).

Officer perspective
Three main themes were identified in relation to how officers perceived and experienced the walkthrough process with autistic detainees: i) the custody context ii) barriers to communication and iii) training and experience (see Table 2). These factors also affected the identification of autistic individuals in police custody.

Restrictions of the custody context
The custody sergeant and detention officer described the demands of the custody environment and their role in custody.

Time pressure
The officers identified competing pressures in their day to day roles which could prevent them from spending more time with detainees than they would wish:

"(…) some -like me -need to spend more time with some people as opposed to others. (…) 'cause of custody and the beast that it is (…) the fact that we may want to spend some time with this particular person, but because, this other person's come in and caused -(…) a bit of a rucus, it -it causes no end of problems. (..″ (Officer interview).
The need to process detainees quickly especially during busier periods was also a factor: "it's a time thing isn't it as well? (…) It's bang, bang, bang, bang, the next person. (…)" (Officer interview).

Lack of alternatives to custody
Another factor which influenced their ability to respond to autistic detainees was the limited alternatives to detention: "Doing this Table 2 Summary of themes for data from officers. job of criminalising people when we don't need to. Arresting them and being in custody when they don't need to be here. And especially if it causes -like, I could see with [Ben] it was causing him that much hassle and grief (…)" (Officer interview). While the custody sergeant referred to the possibility of diversion, he described how this occurred at the end of the process. He also emphasised how it was important for arresting officers to consider the necessity of bringing individuals to custody in the first place.

The impact of custody design on detention processes
Finally, the custody environment itself was considered to play a role in the way officers were able to respond to detainees. In particular, the lack of privacy was highlighted as a factor that might discourage detainees from sharing important medical information: "(…) it's difficult. 'cause sometimes a lot of people hide things. Er, and a lot of people do not want to tell you (…) it's not very private is it? (…)" (Officer interview).

Barriers to communication
Communication between the custody staff and the autistic participants was difficult for both parties, not only for Anne and Ben:

"(…) it was harder with [Anne] than it was with [Ben]. (…) much harder to get the answers and responses from her (…) I've booked a couple of people in who've, um, said they're on the autistic spectrum -(…) and normally get more interaction with them (…) So it's quite difficult… (…) I'd ask them… um… what affects them, and… how do you feel, and how can I make your stay better -(…) and they normally tell us (…)" (Officer interview)
One key barrier to communication was the question structure, particularly with regard to disclosure. This is because there is no flexibility to adjust what is asked according to the needs of the detainee: "(…) all our questions, everything that I went over, are preset." (Officer interview). In particular, the custody sergeant suggested it would help them if there was a set question about autism in the system:

Knowledge and understanding of autism
It was suggested that the response of officers was also influenced by their knowledge and understanding of autism. This was linked to two factors -prior experience with autistic individuals and training.

Prior experience
Having personal experience with autistic individuals was thought to influence officers' responses to autistic detainees: "(…) I only ask certain questions [beyond the pre-set questions] because of my life experiences (…) if I'd not had that experience, I wouldn't ask certain questions. (…) Or I wouldn't be prepared to treat them any different (…)" (Officer interview). It was also observed that prior experience with autistic detainees in their role may improve their responses. During the walkthrough, it was noted that the officers may have adapted their interactions with Ben (the second participant) according to what they learned with Anne (Field notes). In contrast, their overfamiliarity with the role may prevent them from understanding an individual's needs: "I think sometimes we take it for granted (…) the experiences of people coming to custody -especially when they're brand new. (…) we don't quite understand their anxieties. (…) 'cause (…) we see people coming in all the time. (…)" (Officer interview).

Training
It was also implied that the training they receive may play a role. However, they referred to a lack of specific autism training: "We do get a little bit but (…) in email form, really. (…) we do have training -training is good [but] … it's more mental health based (…)" (Officer interview). The custody officer suggested providing specific information to help them support autistic detainees better, such as "quick [reference] cards" (Officer interview).

Discussion
This study was conducted to learn more about the experiences of autistic individuals and the officers interacting with them during the custody process. It aimed to identify potential barriers that might prevent autistic individuals from participating in the custody process and limit the ability of custody staff to support them. The data from the autistic participants showed three overarching themes: i) impact of the custody environment ii) the contribution of communication difficulties to anxiety and iii) impact of sensory sensitivities on autistic individuals' experiences in custody. These themes partly overlapped with those from the officer data: i) restrictions of the custody context ii) barriers to communication and iii) knowledge and understanding of autism. However, the reasons underlying some of the common themes were different. This reflects the differing perspectives of the autistic individuals and the officers who participated in the walkthrough, illustrating how a "double empathy" problem (Milton, 2012;Milton et al., 2018;Sheppard et al., 2016) may present itself in a real-world context.

Custody environment/context
A key issue identified from the data was how the custody environment impacted upon the well-being of the autistic participants. The physical layout of the custody suite increased anxiety due to its unfamiliarity, small/closed-off spaces, bright lights and lack of privacy. Interestingly, lack of privacy was also mentioned in the officer interview but in relation to disclosure of personal information, rather than the impact on detainee wellbeing. The emotional impact of the custody environment, although lessened by the conditions of the walkthrough, was clearly explained as a concern by the autistic participants. Feeling overwhelmed and worried about the consequences of arrest (e.g. losing job) were all highlighted as emotions they would feel quite intensely if they were taken into custody as a suspect. These findings are similar to those made by Skinns' research which discussed the general experiences of non-autistic suspects in police custody (Skinns, 2011). However, the additional sensory sensitivities and communication barriers autistic individuals face are likely to amplify the intensity of these issues, making the custody environment particularly challenging.
The officers highlighted other issues relating to the custody context which they felt restricted their ability to better support autistic individuals in custody. Specifically, time constraints and the lack of available alternatives to detention. These findings are consistent with Crane et al.'s (2016) study which surveyed a large number of police officers and highlighted time pressure and frustration in relation to other legislative restrictions. This emphasises the organisational and procedural constraints operating within police environments and suggests there may be a need for higher level organisational or policy changes.

Communication / anxiety
Communication was perhaps the strongest theme. For the officers, the primary concern was their ability to obtain sufficient information from autistic individuals to help support them. However, the data suggests that the reliance on gathering this information solely through verbal exchange can be a substantial barrier to communication. The key issue for autistic participants was the ambiguity or vagueness of the questions which had implications for disclosure of important medical information or their autism diagnosis. Both the officers and the autistic individuals felt the pre-set questions did not adequately allow individuals to disclose their autism. This was due to the autistic individuals not considering their diagnosis to be an illness, injury, medical condition, learning disability, or mental health condition. While some autistic individuals may have associated learning difficulties (Emerson & Baines, 2010) or mental health conditions (Lai et al., 2019), autism is not classified as either. Furthermore, there has been a movement away from a medical model of autism towards the social model, specifically with regard to understanding autism as another layer of diversity within society (Robertson, 2010). The concept of neurodiversity has gained popularity among some autistic individuals who feel that autism is a part of their identity (see den Houting, 2019 for discussion). This indicates a risk that some autistic individuals may not be identified in police custody. Therefore, the issues in this research with disclosure suggest some changes may be needed at a procedural or policy level to incorporate a question about autism, or at least about needs arising from neurodiversity (i.e. alternative communication support).
An important finding in relation to the theme of communication for the autistic participants was how strongly it was linked with anxiety. Specifically, being informed about what was going to happen was key to helping reduce feelings about uncertainty. It was suggested the addition of visual support (i.e. a toolkit) to convey information in a more accessible way could be valuable.

Sensory sensitivities
Sensory sensitivities were highlighted as having the potential to negatively affect experiences of autistic individuals during the custody process. Tactile sensitivities were emphasised most, particularly in relation to the search and processing procedures. The visual glare from the lighting and glossy colours used in the station also added to their difficulties. Interestingly, smell was not mentioned during the walkthrough. Similarly, auditory sensitivities were not as strongly highlighted. This may have been due to using the training suite rather than the actual custody suite which was always in use. Thus, the impact of these factors may be areas for further investigation.

Knowledge and understanding of autism
The officers highlighted how previous experience with autistic or other vulnerable detainees prompted them to ask additional questions to better support them during the walkthrough. However, although officers have a good awareness and sensitivity to vulnerable detainees, they may be lacking in knowledge of autism specific support. This was clearly illustrated by the quote in section 3.1.2 (above) which conveyed that the officers showed compassion but lacked specific knowledge about how autism affected them. As mentioned by the officers, this may be due to the limited autism training they receive. This is consistent with previous research which raised concerns about the adequacy of autism training for police (Crane et al., 2016).

Recommendations for support
In light of these findings, several recommendations can be made to improve the support of autistic individuals in police custody (see Table 3). These are divided into those that could be implemented by police officers on the ground immediately and those that require a change in policy or practice at a higher level. Notably, many of these recommendations would also benefit suspects more generally or those with other developmental conditions (i.e. sensory processing disorder). Therefore, it would be extremely valuable to consider how some of these changes could be incorporated as part of standard custody procedure, and not only for detainees who disclose a diagnosis of autism. This is particularly important as some individuals with autism may be undiagnosed or fail to disclose they have autism during the booking in process as evidenced in our current findings and in previous work (Crane et al., 2016).
While this study provides valuable insight into the experiences of autistic individuals during the custody process, it also has limitations. The findings are limited to two autistic participants who both had attended University and only required moderate support in their daily lives. Given the heterogeneity of autism, it will be important to replicate this work with other autistic individuals representing a greater diversity of the population. Furthermore, although many of the custody processes will be the same in other police stations (at least in England and Wales), it is important to note that the facilities or environment may differ across police stations. The study is also an example of a 'best case scenario' and various forms of support were in place to help safeguard the welfare of the participants. As the findings do not reflect the heightened stress experienced by detainees in real life, they may not reflect the full range of difficulties some individuals may experience in police custody or the impact this may have on them. Moreover, this study did not include neurotypical individuals. This prevents further comparison of the experiences of autistic individuals with those from the general population. Nonetheless, the autistic participants did report a considerable number of issues that would likely only be amplified under real-world conditions. To gain further insight into these experiences, further research across different settings with a range of participants should be conducted.

Conclusions
The effects of communication barriers and sensory stress illustrated by this study indicate that autistic individuals are at greater risk of being adversely affected when detained in police custody. When combined with the typical difficulties suspects already experience, these additional factors are likely to affect how autistic individuals participate in the custody process. Adaptations to custody procedures and greater provision of accessible information are crucial as a lack of understanding can exacerbate anxiety levels. Taking steps to minimise sensory demands by adapting or better designing custody suites is also important to prevent sensory overload and reduce overall stress. In summary, these findings emphasise the need to implement a range of additional support in police custody and to make changes to the custody environment, police procedures and current policies.

Funding
This work was supported by the Economic and Social Research Council [grant number ES/J500100/1], by a PhD studentship to CH.

Table 3
Recommendations for changes to better support autistic individuals in custody.

Declaration of Competing Interest
None.