Identifying Awareness of Early Offending Behavior in Adolescents with Autism/ADHD

Abstract

The purpose of this study is to explore how adolescents in self-contained classrooms or schools for students with autism or ADHD, with no prior involvement in criminality, perceive and interpret different forms of early offending behavior through fictional case stories. The study specifically aims to examine their ability to discern what constitutes offending behavior, based on the double empathy problem. In total, 13 participants currently receiving secondary-level education (grades 10–12, aged 16–20 years) in self-contained classes at schools for adolescents with autism or ADHD participated. No student had cognitive disabilities or had been involved in any criminal act or criminal justice issues. The students were individually given three fictional written cases of offending behavior (theft, physical assault, and sexual assault). Audio-recorded stimulated recall interviews were obtained while the students solved tasks in relation to the cases, and these were analyzed to capture whether and what aspects of early offending were discerned. Overall, the results indicated limited awareness and enhanced social vulnerability, risking unwitting engagement in early offending behavior. Adapting social science education to students’ special educational needs to understand social interactions might be used to prevent and enhance their awareness of early offending behavior.

1. Introduction

Early offending behaviors in this article refers to “a juvenile’s actions or life conditions that legally allow the juvenile court system to intervene” (Frazier et al., 2024, p. 175). Early offending behavior is operationalized in line with criminological definitions of juvenile delinquency, referring to actions that bring a young person within the jurisdiction of the juvenile court system. For example, youth who have committed an act classified as delinquent under criminal law (i.e., a crime if committed by an adult) constitute early offending behavior (Loeber et al., 2003).

Autism and attention-deficit/hyperactivity disorder (ADHD) are neurodevelopmental disorders that do not inherently increase the likelihood of engaging in criminal behavior (Brewer & Young, 2015). Thus, it is essential to recognize that autism or ADHD does not make someone more likely to be a criminal (Rutten et al., 2017). Hence, it is crucial to approach this topic with sensitivity and avoid generalizing about people with autism and ADHD. Limited abilities in social interaction and awareness might result in adolescents with ADHD and autism being unaware of the implicit intentions others might have in social situations (Gillberg, 2001; Rosello et al., 2020). To further enhance their chances of making positive life choices, early establishment awareness of what constitutes offending behavior among adolescents with ADHD or autism is essential. Neurotypical adolescents usually generalize what they learn in one situation, retain it and use the knowledge in new situations, and learning is refined through social contexts (Takemura, 2021). Different ways of understanding social contexts might be a challenge for people with autism and ADHD, as this involves trying to discern norms and expectations which are implicit in social discourse. Equity in education is based on maximizing the possibility of developing knowledge about how to prevent students with autism or ADHD from being involved in offending behavior. Proper diagnosis, understanding, and educational support can make a significant difference in helping individuals with ADHD and autism navigate social situations and make positive life choices (Couzens et al., 2015).

• The Double Empathy Problem

In this study, the double empathy problem is used as a lens to understand the difficulties people with different perspectives can encounter in the same situations (Milton et al., 2022). The double empathy problem refers to a breakdown in mutual understanding that arises during social interaction between individuals with different experiential and perceptual frameworks, rather than a deficit located solely within the individual with autism. In interactions between people with and without autism, misunderstandings are understood as reciprocal and relational, shaped by differences in social expectations, communication styles, and embodied experiences. This perspective challenges deficit-based accounts of autism by framing social difficulties as an interactional issue rather than an individual impairment. Mainly, the study addresses the risk of people with autism and ADHD being involved in offending behavior without understanding the hidden agendas of other people who are encouraging them to make questionable decisions. Based on this perspective, the double empathy problem also addresses people without a diagnosis and their intentions when using people with autism or ADHD to support offending behavior.

• Autism in relation to offending behavior

Autism is characterized by challenges in social communication and behavior. People with autism may have difficulty understanding social cues and engage in repetitive behaviors (American Psychiatric Association, 2013). Although some people with autism may face challenges in social situations, it is essential to emphasize that autism in individuals does not predispose one to criminal behavior (King & Murphy, 2014). Instead, it is found that people with autism are vulnerable to being victims of crime (Woodbury-Smith, 2024). A study conducted in the U.S. found that adolescents with autism had a high risk of being stopped and questioned by law enforcement (19.5%) in their early twenties (Rava et al., 2017). Factors such as low social motivation, lack of opportunities for interaction, and specific interference responses (i.e., a reduction in social behaviors) may be found suspicious from the perspective of an intendant. It might also comprise challenges which then lead to the development of offending behaviors (Rosello et al., 2020). This finding aligns with results from a review by Collins et al. (2023), which pointed out different motivational factors for offending behavior in the autism group compared to neurotypical adolescents, such as social misunderstandings, idiosyncratic explanations, social naiveté, and circumscribed interests.

People with autism may also have a limited ability to generalize (Erhard et al., 2024); previous experiences do not always help them judge what is expected of them in new situations. Particularly, adolescents with autism can have issues in social contexts due to difficulties in mutual understanding between people who have autism and those who do not have autism (Milton et al., 2022). They may also have a limited ability to inhibit their behavior in social situations (Hlavatá et al., 2018). Therefore, based on the difficulties they experience, some people with autism cannot automatically rely on emotional insights, loyalty, empathy, or compassion to determine whether an action is correct. If a person with autism experiences challenges in relating their behavior to the reactions of others, they might have an impaired ability to realize or recognize what actions may result in inappropriate social behavior (Øzerk et al., 2021). If it becomes difficult to understand what is wrong, it also becomes difficult to understand judgment or punishment, and the risk of recidivism becomes high. Another risk factor of early offending behavior is deficits in empathy and moral reasoning, which might lead to a decreased ability to identify intentions related to eventual criminal activities (Grant et al., 2018).

• ADHD in Relation to Offending Behavior

Adolescents with ADHD may struggle with attention, focus, and impulse control, which can affect their decision-making (Mowinckel et al., 2015). Nevertheless, individuals with ADHD mainly live successful law-abiding lives with or without appropriate support. However, there is strong evidence that individuals with ADHD have a significantly increased risk of early offending behavior (Retz et al., 2021). A national study in New Zealand showed that young adults with ADHD were significantly more likely to be subjected to offending behavior (Anns et al., 2023). However, ADHD is not always diagnosed before a person enters the justice system. A study of prisoners who were screened for ADHD showed that 32.4% met the criteria for ADHD (Capuzzi et al., 2022). In addition, individuals with ADHD might also suffer from the double empathy problem, in relation to executive function deficits (Tatar & Cansız, 2022). Executive functions are cognitive skills that include reasoning, planning, problem-solving, and organizing one’s life (Blair, 2017).

• ADHD and autism in convicted criminals

A 2001 study found that in a group of 126 convicted criminals, 15% had clear signs of autism and an additional 12% had a “high probability” of the condition (Siponmaa et al., 2001). This number contrasts with the 1.08% prevalence in 2021 of the general population in Sweden1. Another study found that 55 out of 100 convicted criminals who underwent forensic investigations had some form of neuropsychiatric disorder (Soderstrom et al., 2005). Persons with autism or ADHD have impairments in their executive capacity (Russell, 1997; Tatar & Cansız, 2022) and, therefore, risk having a weaker ability to organize their lives. As criminality also requires planning and organizing skills, this ability impairment does not explain why these individuals are more frequently found in the justice system than neurotypical individuals. However, another study showed an increased risk of searching for stimulation and risk-taking for adolescents with ADHD (Van Goozen et al., 2007), which may explain somewhat more impulsive than planned criminal acts. Impaired central coherence, including lack of contextual sensitivity or context blindness (Vermeulen, 2015) and enhanced social naiveté, may increase the risk of early offending behavior, e.g., extremism (Rousseau et al., 2023).

According to Mayes (2003), another reason people with ADHD or autism are convicted of crimes is that their impairments result in a decreased ability to participate effectively in hearings, lawsuits, and other situations in which they need to defend themselves. Mayes (2003) and Wing (2013) suggested that people with neuropsychiatric diagnoses are convicted and brought to trial more often than others because of their inability to defend themselves in court. Strategies of self-preservation to advance the rights of human beings are essential, and the ability to explain one’s intentions is vital in such situations. They might also appear suspicious and, therefore, be found guilty, even for crimes they have not committed (Smith, 2022). Wing (2013) described adolescents with autism who joined criminal gangs to enjoy the closest friendships they had ever established. “It is probable that, sooner or later, [they] become the only member[s] of the gang to get caught, while the others are fast enough to get away” (Wing, 2013, p. 201).

1.1. Theoretical Framework

This study examined adolescents’ awareness of early offending behavior using knowledge questions and stimuli recall interviews based on three fictionally written cases of offending behaviors. The study focused on awareness and knowledge of early offending behavior, which are aspects that adolescents with ADHD or autism need to be able to discern to avoid early offending behavior. Hence, variation theory (Holmqvist, 2009; Holmqvist et al., 2014; Marton, 2014) was used to design the cases and analyze what aspects of the cases the participants discerned.

Variation theory is a theory used to analyze aspects of a phenomenon which are discerned and which are not yet discerned in specific situations or phenomena (Holmqvist & Selin, 2019; Marton, 2014). Based on variation theory (Holmqvist et al., 2014; Marton, 2014), this study’s phenomenon is offending behavior, and students’ awareness of what might be early offending behavior in different everyday situations. According to the theoretical framework, discerning aspects of a phenomenon not previously discerned allows for learning, and searching for these aspects offers knowledge of how a powerful learning situation can be designed. Such learning is crucial to create change and understand the difference between delinquent and non-early offending behaviors. Highlighting and directing learners’ attention to undiscerned aspects of a delinquent situation may improve their ability to recognize such behavior and thus avoid participating in future offending behavior. Identifying aspects that define early offending behavior is vital to the ability to learn and change behavior. General knowledge of the critical aspects of specific examples of early offending behavior can help students with autism and ADHD to avoid offending behavior. Using variation theory to define their awareness of these aspects makes it possible to design powerful learning situations preventing early offending behaviors. Furthermore, based on the dual empathy problem, there might be challenges for people with and without autism and ADHD, regarding how each group understands early offending behaviors, and it is important to highlight these issues to prevent vulnerable students from engaging in such activities.

1.2. Aim and Research Questions

As shown above, most students with ADHD or autism do not develop early offending behaviors. Nevertheless, their knowledge of what might be considered offending behaviors can be enhanced to prevent them from engaging in such activities. Guina et al. (2022) highlighted a research gap regarding knowledge about training to prevent people with neurodevelopmental disorders from early offending behavior, despite the group being overrepresented in the criminal justice system. Therefore, to decrease the risk of students with ADHD or autism being involved in early offending behavior, this study aimed to examine early offending behavior awareness in adolescents with ADHD or autism who did not have intellectual cognitive impairments and were not involved in any criminal acts or the criminal justice system, as decreased awareness is assumed as a risk for future early offending behavior. The respondents were adolescents in upper secondary school (grades 10–12; four women, nine boys). This study sought to learn how to prevent people with ADHD or autism from being at a higher risk of engaging in criminal activity than the rest of the population. The research question was as follows:

RQ. 

What early offending behavior awareness is expressed by students with ADHD or autism based on their reasoning about offending behavior portrayed in written fictive stories?

2. Materials and Methods

2.1. Participants

A group of 13 adolescent students with ADHD or autism (aged 16–20 years; three girls, nine boys) in resource classes in general upper secondary school resource units or schools have participated. The selection criteria included students who have symptoms of autism and ADHD, resulting in a need for resource units, without having any intellectual impairments. More specifically, they do not have any problem understanding written text and they all verbally communicated about the text. They have all individually read three fictionally written cases of early offending behavior and answered questions about their understanding of the situations. To avoid handling sensitive personal data, we focused on students placed in resource units for adolescents with autism or ADHD instead of collecting personal data about their diagnoses or health as this was a prerequisite to be accepted as a student in the classes. In this study, the impossibility of carrying out the study in ordinary classrooms was an inclusion criterium for participating in the study. As 98.8% of all students in Sweden are enrolled in mainstream education (European Agency for Inclusive Education, 2015), students placed in resource units have difficulties that are considerable. The teachers have ensured that the students at the group level had formal diagnoses and no involvement in the criminal justice system. There are only a few resource classes for students with ADHD or autism without intellectual disabilities in Sweden, and the two units chosen were part of the two schools in the same region offering such resource classes, both of which only accepted students with autism and ADHD without intellectual disability. The students needed a stimulus-reduced environment to be able to study as they had difficulties with ordinary school environments. Two students declined participation. Thus, the sample comprised the entire body of students we could recruit in this region. As each participant responded to three stories, the dataset consisted of 39 analyzed data elements.

2.2. Procedure

In this study, the participants are reading three different texts which describe theft, physical and sexual assault. They are asked to answer some questions regarding the content to capture their understanding of the written text, as well as how they understand the characters’ reasoning. The first step was to design three fictive written stories of different types of crimes: theft, assault, and sexual assault. When developing the three written cases, aspects of offending behavior were inserted against a background of ordinary daily activities. This offered participants the chance to discern and recall offending behaviors. The stories were inspired by articles in local papers about early offending behaviors that adolescents in the region had been involved in. The fictive stories are presented in Figure 1, Figure 2 and Figure 3.

To capture the students’ awareness, stimulated recall interviews (SRIs) were designed. SRI is a qualitative research method used to access participants’ cognitive processes, perceptions, and decision-making as they occurred during a prior activity. In an SRI, participants are usually exposed to concrete stimuli, most commonly video or audio recordings of their own behavior, shortly after the original event, in order to prompt recall of thoughts, intentions, emotions, and interpretations that were present at the time of action. In such cases, the interview situation is deliberately structured to minimize retrospection bias by anchoring participants’ reflections to observable events rather than generalized post hoc explanations (Lyle, 2003).

In this study, the stimulated recall interview (SRI) was adapted to use fictive written stories as the recall stimulus rather than recordings of participants’ own prior behavior, to avoid sensitive data collection. Participants were presented with short, fictional narratives that they had not encountered previously and were asked to respond to the scenarios in real time. The subsequent interview focused on eliciting participants’ immediate interpretations, reasoning processes, emotional responses, and decision-making as prompted by specific elements of the narratives. This adaptation of the SRI approach enabled the examination of participants’ cognitive and interpretive processes as they unfolded during first-time engagement with a controlled stimulus, while minimizing the influence of memory decay or retrospective rationalization. The written stories functioned as standardized stimuli, allowing for comparability across interviews while preserving the situational and reflective qualities characteristic of stimulated recall methodology.

The questions asked during the interview were designed based on Kaland’s (2003) model to develop an understanding of everyday phenomena. In this study, three fictional cases of offending behavior (stealing, assault, and sexual assault) were the topics in the designed stimuli for the interviews. The three short stories (Figure 1, Figure 2 and Figure 3) were followed by 38 (Case 1–15, Case 2–11, Case 3–12 questions) questions, requiring the respondent to recall and reflect on the content of the stories. The questions followed the same design used by Kaland (2003) for everyday stories supporting daily life activities. The questions were designed to provide respondents with the opportunity to express their awareness of early offending behavior without explicitly describing the offending behavior in the stories.

The interviews were conducted at school, during school hours.

The open-ended questions were divided into three categories: reading comprehension questions (27), physical or environmental (4) and mental or social circumstances (7). Physical inferences (PIs) included facts about a situation such as place and time, whereas mental inferences (MIs) included social interactions or intentions of others. Mo et al. (2019) found that children with autism focus on nonsocial visual peripheral information stimuli more quickly than on social stimuli. Thus, it was hypothesized that students would prefer to answer the A questions as they refer to physical circumstances while B questions refer to mental circumstances. In both cases the RC questions had answers which could be found in the text. On the other hand, the inference questions in each category (PI and MI questions) required the respondents to describe what they considered as offending behavior in each scenario. The texts did not provide explicit definitions or answers, as they did regarding the reading comprehension (RC) questions, requiring participants to rely on their own interpretive judgment rather than textual cues. RC questions with right or wrong answers (27) were used to ensure respondents’ reading comprehension (Case 1–11, Case 2–8 and Case 3–8). This was done to ensure that the participants had understood the content of the text. If they had not understood the content, the analysis of their discernment of offending behavior might be due to poor reading comprehension instead of decreased awareness of early offending behavior. The PI questions (4) were in Case 1–2, Case 2–1 and Case 3–1. The MI questions (7) were in Case 1–2, Case 2–2 and Case 3–3.

2.3. Material and Analysis

The structured SR interviews were audio-recorded and transcribed verbatim. The first step in the analysis was the assessment of the 19 PI questions about the facts in the text to ensure that participants understood its content.

Expressed awareness was analyzed based on variation theory (Holmqvist et al., 2014; Marton, 2014), and data consisted of 11 open-ended questions regarding physical and mental inferences to identify what aspects of early offending behavior were discerned and the respondents’ overall awareness of offending behavior. A qualitative analysis of early offending behavior awareness based on variation theory was conducted. First, the material was read to provide an overview of how the respondents had experienced the cases described. The cases (Figure 1, Figure 2 and Figure 3) included different forms of early offending behavior and the aspects that identified offending behavior were found to be either discerned or not discerned. Otherwise, alternative expressions for understanding the case were identified.

2.3.1. Case 1: Theft

The first case (Figure 1) described a situation in which two women worked together in a restaurant. One woman took money from the cash register and gave it to the other. After reading the text, the participants answered two questions concerning physical and mental circumstances. The story design was based on a variation offered to make it possible to discern the offending behavior (theft). The cash register should have been balanced, but it was not. If money was missing, it could be for reasons other than theft. If this aspect was discerned as a justification for taking money from the cash register, the awareness of offending behavior was not explicit. Another aspect was the missed bus and how one of the women could return home after work. Taking money from the cash register was avoided to make the reasoning more explicit. These two questions provided opportunities for students to make PIs and MIs. Answers to these questions were not found in the text, which encouraged students to draw conclusions based on their offending behavior awareness. Other questions were essential to ensure that respondents understood the story.

2.3.2. Case 2: Physical Assault

The second story (Figure 2) involved two boys who played on the same soccer team. One was a goalkeeper, and the other was the defender. During the game, the defender missed a pass, the opposing team scored, and the first team lost. After the game, the goalkeeper punched the defender who missed the pass in the stomach. The designed variation was the two boys’ different intentions regarding their participation in the team: one was interested in soccer and wanted to become a professional soccer player, and the other did not have such intentions and played soccer more as an activity. Another possible variation to discern was the motive for the goalkeeper to hit the defender, as the latter had made a mistake, which might make students justify the offending behavior in the same manner as in the case of theft, where Lotta did not take the money from the cash register and so, was not considered to have any debt. Based on this, there was one PI question asking why the team lost the game and two MI questions focusing on why the goalkeeper hit the defender and why he was afraid of missing the opportunity to become a professional soccer player.

2.3.3. Case 3: Sexual Assault

In the third and final story (Figure 3), a boy interested in a girl tried to attract her attention in a way he had seen in a movie: by touching her. However, in the movie, the woman probably knew the man, but the girl in this story did not know the boy well and felt insulted when he touched her. The variation used in this case was made to discern the differences in physical contact between people who know and do not know each other. In addition to questions about the story’s content, one PI question and three MI questions were formulated to ensure that respondents understood the case. The PI question required respondents to figure out the day the boy contacted the girl, which was not explicitly stated in the text. The intention was to force their focus on the meeting. The MI questions were directed toward the girls’ reactions to the boy’s behavior to make the respondents conclude why the girl reacted as she did, trying to capture whether they could imagine her reaction and feelings.

2.4. Ethical Considerations

This study adhered to the ethical guidelines of the Swedish Research Council (2017) and those established in the year of the data collection (2009). The participants belonged to two resource units for adolescents with ADHD or autism. All parties involved (including parents, teachers, and adolescents) were repeatedly informed about the study’s aims, use of the data, and their right to confidentiality and withdrawal from participation at any time. The participants and their parents also provided signed consent to participate in the study and for the researchers’ use of the information elicited in the interviews. As the data collected referred to fictive cases in educational settings, not representing the students’ personal situations, and included no personal data about the students’ diagnoses or health, their own experiences or personal circumstances, no institutional ethical approval was obtained. Data were transcribed and any personal or identifying items were replaced with fictive names or deleted. The timespan between data collection and publication reduced the risk of identifying any specific person. Questions addressing scenarios involving sexual assault (in this case caressing a woman’s back) could be perceived as sensitive and could be experienced as emotionally challenging by some participants. At the same time, the situation presented in this study may also be perceived by participants as emotionally distant or non-sensitive, as it did not involve their own lived experiences. Rather, participants were asked to respond to fictional written material, allowing engagement with the topic at a reflective and hypothetical level. This design was intended to reduce the risk of personal distress while still enabling the examination of participants’ interpretations and reasoning in relation to sensitive subject matter.

3. Results

Initially, the students’ answers for the 27 RC questions with right or wrong answers were summarized from all cases and reported descriptively, divided by the cases and questions focusing on physical or mental circumstances (Case 1–11, Case 2–8, and Case 3–8) (

Table 1. Reading comprehension (right/wrong answers). P—physical. M—mental aspects.

Case	Aspect (27)	Total	Women	Men
Theft	P (6)	43 (54) 80%	19 (24) 79%	62 (78) 80%
	M (5)	37 (45) 82%	14 (20) 70%	51 (65) 78%
Assault	P (4)	26 (36) 72%	10 (16) 62%	36 (52) 69%
	M (4)	39 (45) 87%	15 (20) 75%	54 (65) 83%
Sexual assault	P (5)	39 (45) 87%	17 (20) 85%	56 (65) 86%
	M (3)	26 (27) 96%	10 (12) 83%	36 (39) 92%

). The results show a high level of reading comprehension. The case with lowest results is the football situation (assault). A case-related analysis of the RC questions is presented below.

3.1. Theft

The results show that it was easy for participants to answer questions presented as facts in the text (

Table 2. Correct answers: physical aspects of Case 1 (theft).

Question	Boys (n = 9)	Girls (n = 4)
A1	9	4
A2	3	2
A3	9	4
A4	9	4
A6	6	3
A7	7	2

and

Table 3. Correct answers: mental aspects of Case 1 (theft).

Question	Boys (n = 9)	Girls (n = 4)
B1	6	0
B2	9	4
B4	8	4
B5	5	3
B6	9	3

). One question, inquiring about the season, only received a few correct answers, as the information may have been too peripheral (5 out of 13 points).

For questions concerning interactions and people, some respondents required clarification of the characters’ names, resulting in incorrect answers. The respondents generally required help with questions that required inferences based on the students’ conclusions. Question A5 (PI) asked why daily receipts and cash were seldom balanced and question A8 (PI) asked why Lotta could not go home by bus. Three students answered question A5 with more than one reason for the cash imbalance, suggesting staff, computer systems, or other variables as the cause.

However, most respondents referred to the specific text, such as the cash register needing to be balanced because Annika used to take money from it to go home by taxi. Question A8 was answered by all respondents for only one reason: the bus had already left. No comments were made about Lotta being late because the guests stayed too long. Comments were made about how Annika took the money and that Lotta was innocent because she did not take the money out of the cash register. The results showed that these adolescents had difficulty understanding different perspectives and had poor judgment of probability. Questions referring to MIs seemed even more difficult. Question B3 asked why Lotta was reluctant to take the money and question B8 asked why she felt guilty the following day. The respondents required help in predicting and understanding other people’s motivations and reactions. The answers to question B3 varied between “it was stolen” and “she wanted the daily cash to be balanced,” and those for question B8 varied between “the money was stolen” and “it was not right.” Only one respondent made other interferences: “You cannot feel guilty if you do not know it is stolen, but in this case, she knew, and so she felt guilty even if she did not take the money herself.” The answers presented were simplified and contextualized, representing a single perspective; however, the awareness of this act as offending behavior seemed to be discerned by most respondents.

3.2. Assault

The results of the assault case aligned with those of the first case (

Table 4. Correct answers: physical aspects of Case 2 (physical assault).

Question	Boys (n = 9)	Girls (n = 4)
A1	3	1
A2	7	2
A3	8	3
A4	8	4

and

Table 5. Correct answers: mental aspects of Case 2 (physical assault).

Question	Boys (n = 9)	Girls (n = 4)
B1	9	3
B2	7	3
B3	7	3
B4	7	2
B5	9	4

). Only the first question with physical or factual focus, on how the boys knew each other, appeared difficult to discern. Even the mental and social questions seemed easy to understand, and the main error was mixing up the boys’ names or misremembering.

Vedran was talking about seeing a movie after the match, rather than the game. There were three inference questions: one for PI and two for MI. Question A5 asked why the team lost the game. The text described how Vedran made a wrong pass. Although one single pass would not likely be the only reason for losing an entire game, 12 out of 13 participants answered this question with “Vedran made a wrong pass, and that is why they lost,” or “a wrong pass.” No one reflected on the question from any other perspective, such as suggesting that the team not scoring or having a poor defense could have been the reason for the loss. Responses to the two MI questions (question B6, asking why Niklas hit Vedran and question B7, asking why Niklas thought his dream of becoming a professional player might be in danger), showed almost the same pattern. Question B6 was answered in one of two distinct ways by most respondents: either that it was Vedran’s fault that the team lost (6) or that Niklas was angry (6). However, for question B7, two respondents gave answers that considered different possibilities: one suggested that Niklas was angry at Vedran for the failed pass, and another proposed that the goalkeeper might have been disappointed in himself for allowing the goal. The analysis did not provide indications of awareness of hitting another person as an offending behavior.

3.3. Sexual Assault

Following the results of the previous two cases, there were no difficulties related to what happened in the story (facts). The respondents missed answering some of the questions as they did not discern the answer and claimed that some information was “not being in the text.” It was easy for the respondents to understand what happened between the people in the film, and almost all the respondents recalled and answered any related questions correctly (

Table 6. Correct answers: physical aspects of Case 3 (sexual assault).

Question	Boys (n = 9)	Girls (n = 4)
A1	8	4
A2	7	3
A3	7	4
A4	9	4
A5	8	2

and

Table 7. Correct answers: mental aspects of Case 3 (sexual assault).

Question	Boys (n = 9)	Girls (n = 4)
B1	8	4
B2	9	3
B3	9	3

).

In this case, one PI question was asked about the day Samuel met Hannah after watching the movie (question A6). The text stated that they met the day after Samuel had watched the movie, and that Samuel met Jonas on a Thursday. Six respondents made inferences to find the correct answer. The rest said that they did not know or “it was not in the text.” Finally, three MI questions were asked: B4 (why did the woman in the movie act as she did?), B5 (what did Samuel think would happen?), and B6 (why was Hannah angry?).

Again, very few answers reflected the consideration of different perspectives. Only one respondent answered question B4 with two alternative reasons: either that the actress’ behavior was part of her role or that she liked the man. The common answer was that she acted the way she did because it was a movie, and by implication, such behavior would not occur in real life. Most respondents answered question B5 by saying that Samuel thought Hannah would react as the woman in the movie did. Only one person said that Samuel hoped to be part of a couple and thought that this attempt at contact would make this happen. Question B6 was answered in several ways but with only one solution per respondent. Six answers were related to the opinion that Hannah did not like being touched by Samuel and was angry at him for his presumption. The analysis showed that the respondents did not discern what was wrong and how it might be an early offending behavior, as touching a person’s private parts without consent is defined as an offending behavior according to Swedish law. Even if the questions were based on Hannah’s perspective, no one noticed that Hannah had the right to be angry as an offending behavior had been conducted against her.

4. Discussion

This study aimed to investigate early offending behavior awareness expressed by upper secondary students placed in self-contained classrooms or schools for students with autism and ADHD. As Rosello et al. (2020) pointed out, a lack of opportunities for interaction and reduced social behavior may hinder the development of offending behavior awareness. The results presented above regarding offending behavior showed some differences. In the case of theft, the respondents discerned that it was wrong and discerned the aspects of early offending behavior. The other two cases (assault and sexual assault) were more complicated, and the situations leading to the early offending behavior were more in focus than the behavior itself. Hurting one physically or sexually seemed to be more difficult to discern as offending behaviors, which might be related to double empathy challenges. Theft is physical, relating to things that belong to others and should not be taken away. Feelings, pain, and hurt are phenomena that are implicit and abstract and require a developed sense of double empathy. In line with Collins et al. (2023), these results show that the students focused on aspects other than the offending behavior; for example, justifying the theft by saying that the woman who used the money did not have any guilt as it was the other women who took the money, or that the girl liked the boy, and by extension, appreciated being touched even if it was by an unfamiliar boy. The research question addressed early offending behavior awareness expressed by the students concerning the presented cases, and the results point out that awareness appeared to be limited regarding acts that could not be visually understood, unlike physical objects being stolen. Capturing and describing the perspectives of adolescents’ understanding of offending behavior regarding assault is essential to help prevent them from engaging in early offending behavior. The present study focused on adolescents without engagement in criminal behavior, striving to prevent a vulnerable group from such development by enhancing their own understanding of what can be an offending behavior and by guiding them to avoid such situations. Collins et al. (2023) showed that the motives for criminal acts by people with ADHD or autism are different in comparison with neurotypical people, and this study is a humble step toward understanding the various reasons for early offending behavior.

Limitations

This study had several limitations. First, its sample size was small, and participants were placed in segregated educational contexts with a high number of adults and small groups of students; therefore, they probably did not have any opportunities to engage in early offending behavior, at least not during school time. Secondly, the study was conducted several years ago, when fewer adolescents were involved in serious criminality (The National Audit Office, 2024). Finally, as no personal data were collected on the respondents’ health or symptoms, there might have been differences between individuals owing to their different personal symptoms.

5. Conclusions

Based on the variation theory, knowledge is assumed to rely on the ability to see a whole and its parts simultaneously, while varying those parts and still relating them to the whole. In this case, 13 respondents answered questions regarding three cases, providing a data sample of 38 items. The respondents were upper secondary school students with ADHD or autism but without cognitive disorders. They answered questions correctly to a high degree when answers were easily found in the text. However, the results showed that they encountered difficulties in shifting perspectives between the people presented in the same case, which is a very important skill for considering the possible consequences before performing an action. One important finding was shown in the assault case, where challenges related to the double empathy problem were discerned. Only one respondent reflected on the possibility that the team’s loss might not have been Vedran’s fault. Difficulties in discerning different perspectives based on which person’s views are being perceived, and supporting students when they experience such differences in social situations might give them a better understanding of what constitutes offending behavior. The respondents provided correct answers to questions when a template was offered within the stories. Furthermore, their reading comprehension was high which indicate their literal understanding of the situations described. However, it was challenging to discern aspects of offending behavior in a complex context, as well as draw conclusions independently. In real life, there is no template and things happen fast. To support students with autism and ADHD at school, providing definitions of offending behavior seems to be important. To ensure that everyone is provided with the best opportunities, students need to be allowed to discern and reflect on offending behavior in a secure and supportive school environment to prevent everyone from engaging in such activities.