Lay person’s and psychology officers’ beliefs about memory, investigative interviewing and deception detection: data from Malaysia

ABSTRACT

It is very typical for people to think that the understanding of psychological phenomena is a matter of common sense. In many parts of the world, the general public and even some professional groups, often express commonly held beliefs about topics in psychology that are not in line with up-to-date scientific literature. To date, comparable data from Malaysia is lacking. In this study, 101 psychology officers serving the Ministry of Home Affairs as well as 130 student and community members in Malaysia completed a survey on beliefs about human memory, detecting lies and police interviews. Generally, participants possessed beliefs that were not in line with current psychological research, which is consistent with findings from many parts of the world. This study serves as an important gauge of the advancement of the field in this region.

KEYWORDS:

• Memory
• investigative interviewing
• deception detection
• beliefs
• psychology officers

Introduction

Decisions about policies and laws in the justice system are often based on assumptions about human behaviour – the very subject matter of psychology. Scholars have spent decades applying psychological perspectives in the legal system, emphasising the importance of scientific, research-informed practice. Nevertheless, the understanding of psychological phenomena is frequently assumed to be a matter of common sense. Many psychological misconceptions or false beliefs are based on widely held ‘common sense’ beliefs, and Lilienfeld et al. (2009) have argued that such misconceptions impede the development and dissemination of accurate knowledge, as well as lead to poor decision-making in every domain of our daily lives. Addressing psychological misconceptions is therefore an important agenda in the psychology field.

The justice system has profound implications on individual liberties, communities and society as a whole. As an example, misconceptions about criminal behaviour may lead law enforcement agencies to rely on inaccurate profiling techniques, resulting in wrongful suspicion or conviction. Given the high-stakes nature, it is particularly important that the professionals within it are equipped with beliefs or knowledge that is in accord with up-to-date scientific evidence, because such knowledge guides policymakers’ decisions on criminal justice issues. In the current study, issues around people’s beliefs about memory, police interviewing and lie detection were explored.

Beliefs about memory

Early literature has indicated that people often have certain beliefs about how memory works. For example, Loftus and Loftus (1980) reported findings from an informal survey that the majority of respondents believed every detail we learn is permanently stored in memory, albeit not necessarily accessible at all times. Several studies since then have continued to find that many subscribe to the belief of memory permanence (e.g. Simons & Chabris, 2011; Wake et al., 2020; see also Brewin et al. (2019) for a critique on question wording on belief surveys). However, years of research evidence show that the human memory does not work like a video recorder, and that it is malleable, easily distorted by misleading information, and is reconstructive in nature (Lacy & Stark, 2013).

Studies from different countries have reported that even professionals exhibit a lack of specialised knowledge about memory. In Norway, Magnussen and Melinder (2012) surveyed over 800 licensed psychologists about their knowledge of memory science and found that they did not perform significantly better than lay people or trial judges. For instance, evidence from memory research indicates that children generally remember events worse than adults, but almost half of the psychologist respondents in the study did not think so. In addition, among undergraduate students, the public and psychologists, Patihis et al. (2014) found that most practitioners, especially those who were not research-focussed, were more likely to agree that memories of traumatic events can be repressed and recovered through therapy – a concept that has not been substantiated. In a study by Chung et al. (2022), Malaysian victim care officers who typically have a background in psychology reported to believe that it is impossible for children to develop false memories of sexual abuse, despite evidence showing that false memories can indeed be implanted through suggestive questioning techniques (Schreiber et al., 2006).

That said, in the United Kingdom, Ost et al. (2017) found that clinical psychologists who receive extensive training showed more accurate knowledge about memory compared with hypnotherapists and undergraduate psychology students. This may indicate that expert training and a commitment to engage in the scientific literature can ensure that the professional’s practice is evidence-informed.

Beliefs about police interviewing and interrogation

Similarly, while interviewing, interrogation and confessions are grounded in fundamental principles of psychology, many of these principles are not well understood by the general public. As shown by Blandón-Gitlin et al. (2011), jurors tend to think that investigators are unlikely to manipulate innocent people into confessing, unless highly coercive tactics such as actual or threats of violence are used in interrogation. Most people are also reluctant to believe that they themselves would falsely confess and therefore evaluate others accordingly (Henkel et al., 2008). However, it has been repeatedly shown through field and laboratory work that, through psychological manipulation, people can indeed be led to confess to crimes they did not commit (for a review see Kassin, 2012). Innocent people are also more likely than guilty people to subscribe to the naïve belief that the truth will prevail and therefore waive their right to remain silent, very often to their own detriment (Kassin & Norwick, 2004) – but members of the public are often not aware of this phenomenon (Kassin et al., 2018).

Kassin’s (2012) review also highlighted that despite being unreliable, confession evidence is typically seen as the gold standard, given that they are usually detailed accounts of a crime which indicates truth-telling, and the common-sense notion that most people would never voluntarily confess to a crime they did not commit. Nonetheless, confessions, even if false, have significant influence on the police investigation process. An experiment by Hasel and Kassin (2009) illustrated how confessions can taint perceptions of eyewitnesses: participants witnessed a mock theft and then made an identification from a target-absent photographic lineup. Upon receiving feedback that another suspect (as opposed to the one identified by the participant) had confessed, a majority of the participants changed their initial identifications and selected the suspect who had allegedly confessed.

Taken together, these findings show that the average layperson and the untrained professional within the criminal justice system may not necessarily understand the link between interrogation tactics and false confession, and its consequences for investigations and prosecutions. The literature also highlights the fact that the current systems in place may not provide adequate safeguard against unreliable confessions. An inaccurate understanding of the psychology of interviewing and confessions impedes improvement to the system.

Beliefs about deception detection

In the controversial Reid’s interrogation manual, Inbau et al. (2001; see also Inbau et al., 2013) have argued that police can be effectively trained to differentiate if a suspect is telling the truth or lying based on nonverbal cues, yet a summary of studies by Vrij (2008) highlighted that while training can enhance accuracy, the effects are minimal. It is also generally assumed by law enforcers that liars are more anxious than truth-tellers, as shown by the dominant use of anxiety-based lie detection tests as well as the abundance of courses promoting lie detection competence.

In an early study by Akehurst et al. (1996), police officers and the lay public were asked to review a list of nonverbal behaviours, and to rate whether they believed these behaviours would increase, decrease, or remain unchanged during lying. While it was expected that police officers – whose work involves judgements of honesty in high-stakes contexts – would be more accurate than lay people in their beliefs regarding deceptive behaviour, this was not the case. For instance, both samples were equally likely to believe that hand and leg movements increase during deception, while they have actually been shown to decrease (DePaulo et al., 1985). In a more recent study, Delmas et al. (2019) presented a set of photos of facial expressions to police officers experienced in conducting interrogations as well as civilians in France. Participants were tasked to determine whether an expression was more or less present during lying. It was found that both police officers and civilians held similar stereotyped beliefs as to what behaviours are indicative of lying, for example, gaze aversion. However, it is normal for anyone accused of crime to display strong psychological states such as fear and anxiety, hence rendering such cues to be unreliable indicators of deceit (for a review see Vrij & Granhag, 2012). The overconfidence in the reliability of such cues can lead to misjudgements of one’s guilt or innocence, as people’s accuracy in distinguishing truth and lies based on nonverbal behaviours is often only at the chance level (Bond & Depaulo, 2006).

Verbal cues, on the other hand, are perhaps better indicators of deceit, as research has explored ways to amplify the difference between what liars and truth-tellers say (Vrij, 2019). For instance, in the Strategic Use of Evidence (SUE; Granhag & Hartwig, 2015) approach, the strategic disclosure of evidence enables guilty suspects to contradict the evidence or their previous statements, which can be an indicator of lying. The Verifiability Approach (VA; Nahari, 2018) of interviewing is based on the idea that liars are tempted to add details when providing their accounts to seem credible, but such details are generally unverifiable. These are just two example approaches that underline the importance of shifting the attention from observing nonverbal behaviour to analysing verbal content, but many in the field are still reluctant to drop the nonverbal lie detection line of research (Vrij et al., 2023). Brennen and Magnussen (2020) rightfully pointed out the importance of debunking pseudoscience so as to ensure guidelines that are recommended to the criminal justice system are credible.

A point to note is that while it is important to be up to date with the current literature, researchers and practitioners alike first need to be confident with the credibility research conducted. Luke (2019) raised concern that suboptimal research practices in the field of deception may have lowered the informational value of the present literature, arguing that current practical recommendations may have been based on cues which are overestimates.

The present study

In sum, it appears that the public and professionals hold beliefs that are not in line with academics who are usually immersed in their respective research expertise. As illustrated in the literature, inaccurate beliefs, or a lack of understanding about applied psychology topics can lead to serious miscarriages of justice.

While many studies have been conducted in various parts of the world, comparable data from Malaysia is currently lacking. In the present study, a survey was conducted among psychology officers serving the Ministry of Home Affairs, students, and community members in Malaysia, in order to examine their beliefs in three areas of applied forensic psychology topics, namely human memory, police interviews and lie detection.

The beliefs of the participant sample under investigation are of particular interest because of their educational background and qualifications, as well as the landscape of the psychology profession in Malaysia. The ‘psychology officer’ position was set up in 2008 under the Public Services Commission of Malaysia, but officers taking on the position may have come from a wide range of backgrounds, including psychology, counselling, or social work, hence the job title ‘psychology officer’ can be seen as a misnomer. There are psychology officers who serve ministries other than the Ministry of Home Affairs, with expertise such as education, human resources, health, among others. Now, the minimum educational qualification to be employed as a psychology officer in the Malaysian government is a bachelor’s degree in psychology or counselling, and the job scope includes policy planning and governance, consultation, assessment, treatment, research, training and public engagement (see published document by the Public Service Department, n.d.a). On the other hand, an assistant psychology officer supports the activities carried out by psychology officers as stated above, including administrative duties (see published document by the Public Service Department, n.d.b). To be employed as an assistant psychology officer, one only needs to have at least a diploma in the field of psychology, counselling, human resource development, or cognitive science.

Such an employment practice in the public service indicates that a considerable proportion of civil servants may not be sufficiently trained in the area of psychology to undertake tasks a typical qualified psychologist with postgraduate training is expected to take, such as administering psychological assessments, making mental health diagnoses, conducting interventions, serving as expert witnesses at court, among others. This situation partly stems from the government’s lack of understanding of what psychological science is and what psychologists are trained to do. Therefore, whether this group of officers hold accurate or inaccurate beliefs about important psychology topics has major implications toward how psychology professionals are regulated in Malaysia.

At present, Malaysia has no laws regulating the psychology profession, even though it is obvious that having certain formal qualifications or experience to practise as a professional psychologist is crucial. The lack of legislation makes it challenging for action to be taken against individuals who falsely claim to have qualifications that meet standards of practice or codes of conduct. In these circumstances, any individual with an educational background in psychology or a cognate area can profess to be a qualified psychologist and undertake professional activities with little to no consequences. Therefore, a study that surveys the beliefs about important psychology topics among those who have studied psychology is particularly useful to gauge the advancement of the field in this region; such data can urge policymakers in Malaysia to establish the necessary safeguards for those with whom psychologists interact professionally.

Method

Participants and procedure

This study was not preregistered, but the data and materials are freely available via https://osf.io/36w9y. The study was reviewed and approved by the University of Reading Malaysia Research Ethics Committee (reference number UoRM REC 2021/04). The first author was invited to deliver an online forensic psychology training session in August 2021, which had 149 attendees. Prior to the training session, the questionnaires were administered via the Jisc Online Surveys platform to psychology officers serving the Ministry of Home Affairs Malaysia. One hundred and one participants (52 males, 49 females) aged 29–55 (M = 36.57, SD = 4.22) provided informed consent and completed the questionnaires. Among the participants who have completed the questionnaire, most (n = 74. 74.3%) served as psychology officers, 24 (23.8%) served as assistant psychology officers, one (1.0%) served as the chief inspector of prison and one (1.0%) did not provide any data. The majority of the participants were ethnic Malay (n = 72, 71.3%), followed by Chinese (n = 9, 8.9%), Indian (n = 6, 5.9%), other ethnicities (n = 11, 10.9%) and three (3.0%) preferred not to report. Almost half (n = 46, 45.5%) had completed at least a bachelor’s degree, 35 (34.7%) had a master’s, 17 (16.8%) had a diploma, two (2.0%) had a doctorate, and only one (1.0%) had secondary education as their highest qualification.

Additionally, the questionnaire was also administered to university students and members of the public, recruited through the [University] [School] research participant pool management system and researchers’ social media networks. There was a total of 205 responses, but those that were incomplete (n = 2) or completed in less than two minutes (n = 9)¹ were excluded. Out of the remaining data from 196 participants, 130 (66.3%) were Malaysian nationals who have lived in Malaysia for at least 15 years and 66 (33.7%) were non-Malaysians.² Given that the current study aimed to gauge the beliefs about applied psychology topics in Malaysia, only data from Malaysians were analysed. The final sample comprised 25 males and 99 females, plus six participants who preferred not to report, aged 18–61 (M = 23.72, SD = 7.79). Most of the participants in the final sample were ethnic Chinese (n = 100, 76.9%), followed by Malay (n = 12, 9.2%), Indian (n = 11, 8.5%), other ethnicities (n = 4, 3.1%) and three (2.3%) did not report. More than half (n = 79, 60.8%) had a bachelor’s degree, 31 (23.8%) had further education (e.g. pre-university qualification such as A-levels) or diplomas, nine (6.9%) had a master’s, eight (6.2%) had secondary education as their highest qualification, two (1.5%) had other professional training, and only one (0.8%) had no formal education. At the time of the study, 52 (41.5%) did not have any formal education in psychology, counselling, or cognitive neuroscience, whereas the remaining had exposure to these areas; specifically, 53 (40.8%) were pursuing studies at university, 21 (16.2%) held a degree, one (0.8%) held a postgraduate degree and one (0.8%) held a diploma.

For data analyses, all 231 participants were categorised into three groups: psychology officers (n = 101), participants with psychology background (n = 76) and lay participants without psychology background (n = 54).

Materials

In the first section, participants reported basic demographic information. Psychology officers also reported the role they held in the Ministry of Home Affairs and the years of experience they have had as psychology officers. All respondents were asked to describe how frequently they read journal articles relevant to research on memory, police interviewing and interrogation, as well as lie detection (1 = rarely, 2 = occasionally, or 3 = regularly).

In the second section, participants completed a memory beliefs questionnaire developed by Ost et al. (2017), which contains 10 statements relating to memory functioning. Participants indicated their level of agreement on a 4-point scale (strongly disagree, disagree, agree, strongly agree), or they could opt for the ‘I don’t know’ option. Prior to analyses, nine of the ten items were first reverse-scored so that higher scores reflected responses that were more in line with current scientific consensus. Responses of ‘strongly agree’ and ‘agree’ were coded as ‘accurate’ beliefs about the statements, whereas ‘strongly disagree’ and ‘disagree’ were coded as ‘inaccurate’ beliefs. The percentage of ‘accurate’ responses for each participant was obtained.

In the third section, participants completed a questionnaire consisting of 30 confession-relevant statements. This questionnaire was developed by Alceste et al. (2021). The scale used is identical to that in the second section. Prior to analyses, eleven of the thirty items were first reverse-scored so that higher scores reflected responses that were more in line with current scientific consensus. Data for the third section were coded the same way as those in the second section.

The fourth section comprised one open-ended question: What do you think are good cues for detecting lies? (following Bogaard et al., 2016). Participants were given space to type in their responses.

Results

Participants were asked to report their frequencies of reading scientific papers about the three applied psychology topics. In terms of memory research literature, the average rating for the Ministry of Home Affairs psychology officers was 1.64 (SD = .72), 1.66 (SD = .62) for people with psychology background and 1.35 (SD = .52) for the lay participants without psychology background. When asked about the interviewing and interrogation research literature, psychology officers reported an average rating of 1.59 (SD = .73), 1.41 (SD = .57) for people with psychology background, while lay participants had an average rating of 1.33 (SD = .51). In terms of lie detection research literature, psychology officers reported an average rating of 1.50 (SD = .72), 1.32 (SD = .55) for people with psychology background, and 1.30 (SD = .50) for lay participants. A summary of these findings is presented in Table 1.

Table 1. Means and standard deviations of self-reported frequencies of reading journal articles about memory, investigative interviewing and lie detection.

Applied forensic psychology topics	Psychology officers (n = 100)	Respondents with psychology background (n = 76)	Lay people (n = 54)	Total (N = 230)
Memory	1.64 (.72)	1.66 (.62)	1.35 (.52)	1.58 (.66)
Investigative interviewing	1.59 (.73)	1.41 (.57)	1.33 (.51)	1.47 (.64)
Lie detection	1.50 (.72)	1.32 (.55)	1.30 (.50)	1.39 (.62)

Note: Items are reported on a three-point scale (1 = rarely, 2 = occasionally, 3 = regularly).

Three one-way analyses of variance were conducted to determine if the three groups differed in how frequently they read scientific articles in each topic area. Specifically, there were differences in how much participants reported they engaged with the memory literature, F(2, 227) = 4.36, p = .014, partial η² = .04, 95% CI [.002 to .09], as well as how much they engaged with the police interviewing literature, F(2, 226) = 3. 36, p = .036, partial η² = .03, 95% CI [.00 to .08]. Tukey post hoc analysis showed that both the psychology officer and participants with psychology background groups reported that they engaged with the memory literature more than those without psychology background, with a difference of .29, 95% CI [.03 to .55], p = .024, Cohen’s d = 0.44, 95% CI [.10 to .77] and .31, 95% CI [.04 to. 58], p = .022, Cohen’s d = 0.53, 95% CI [.17 to .88] respectively. The psychology officer group reported that they engaged with the interview literature more than those without psychology background, with a difference of .26, 95% CI [.00 to .51], p = .045, Cohen’s d = 0.39, 95% CI [.05 to .72].

Beliefs about memory

The percentages of participants who had beliefs in line with current memory research are presented in Table 2. The mean percentage scores were 29.90% (SD = 18.52), 57.11% (SD = 21.47), 40.19% (SD = 22.53) for psychology officers, people with psychology background and lay people with no psychology background respectively. A one-way analysis of variance showed that there were significant differences in performance across the three groups, Welch’s F(2, 124.43) =  38.92, p < .001, η² = .25. Games-Howell post hoc analysis revealed that psychology officers had significantly lower scores compared to those with a psychology background, a difference of 27.20, 95% CI [19.92–34.49], p < .001, Cohen’s d = 1.37, 95% CI [1.04–1.70]. Psychology officers also had lower scores than those without psychology background, with a difference of 10.28, 95% CI [1.76–18.81], p = .009, Cohen’s d = 0.51, 95% CI [.18 to .85]. Participants with no psychology background also had lower scores than their counterparts with some psychology background, with a difference of 16.92, 95% CI [7.58–26.26], p < .001, Cohen’s d = 0.77, 95% CI [.41–1.13].

Table 2. Percentage of participants who had ‘accurate’ beliefs about the statements.

	Statements	Psychology officers (n = 101)	Respondents with psychology background (n = 76)	Lay people (n = 54)	Total (N = 231)	χ^2
1	The mind is capable of unconsciously ‘blocking out’ memories of traumatic events.	11.9	6.6	13.0	10.4	9.44
2	Memory is like a computer, accurately recording events as they actually occurred.	16.8	68.4	37.0	38.5	39.63*
3	It is possible for an individual to develop ‘false’ memories of non-traumatic events.*	78.2	93.4	87.0	85.3	13.77*
4	Very vivid memories are more likely to be accurate than vague memories.	11.9	35.5	16.7	20.8	9.78
5	A poor memory for childhood events is indicative of a traumatic childhood.	31.7	53.9	44.4	42.0	10.23
6	The more confidence with which a memory is reported, the more likely it is to be accurate.	19.8	64.5	27.8	36.4	37.89*
7	Early memories, from the first year of life, are accurately stored and retrievable.	35.6	76.3	50.0	52.4	35.23*
8	Memory is not influenced by suggestion.	52.5	77.6	57.4	61.9	32.44*
9	It is possible for an individual to distinguish between ‘true’ and ‘false’ memories.	20.8	51.3	35.2	34.2	17.12*
10	The more emotion with which a memory is reported, the more likely it is to be accurate.	19.8	43.4	33.3	30.7	12.36*

Note: Item 3 marked with asterisk indicates that it is in line with current scientific consensus. *p < .005 (corrected for multiple testing).

Kruskal–Wallis H tests were also conducted to determine whether there were differences in endorsement scores for each statement among the three groups of participants. Median endorsement scores were statistically significantly different between groups for a number of beliefs about memory, namely Statements 2, 3, 6, 7, 8, 9 and 10. The χ² statistics for each statement are reported in Table 2. P-values have been adjusted to .005 using Bonferroni corrections; this was done by dividing the alpha level of .05 by 10 (i.e. the number of tests). Post hoc analyses showed a pattern where participants with psychology background endorsed statements that reflect the state-of-the-art.

Beliefs about police interviewing and interrogation

Table 3 displays the percentages of participants who provided answers that are in line with current police interviewing research. The mean scores were 57.39% (SD = 13.39), 60.48% (SD = 13.97), 54.57% (SD = 17.10) for psychology officers, people with psychology background and lay people with no psychology background respectively. A one-way analysis of variance showed that there were no significant differences in performance across the three groups, F(2, 228) =  2.67, p = .07.

Table 3. Percentage of participants who had ‘accurate’ beliefs about the statements.

	Statement	Psychology officers (n = 101)	Respondents with psychology background (n = 76)	Lay people (n = 54)	Total (N = 231)	χ^2
1	In general, people are highly accurate judges of truth and deception.	27.7	61.8	51.9	44.6	23.72*
2	Compared to most adults, adolescents exhibit ‘immaturity of judgement’ in their decision-making.*	85.1	75.0	55.6	74.9	13.70*
3	Telling a suspect, falsely, that evidence links him/her to the crime will increase the risk that an innocent person would confess to a crime he/she did not commit.*	65.3	57.9	53.7	60.2	1.15
4	Compared to guilty suspects, innocent ones are more likely to waive their right to remain silent and to have a lawyer present.*	65.3	44.7	46.3	54.1	5.20
5	In general, people’s behaviour is strongly influenced by their expectations of reward and punishment.*	85.1	81.6	70.4	80.5	4.16
6	Trained police can distinguish between truths and lies at high levels of accuracy.	23.8	39.5	46.3	34.2	12.12
7	Sleep deprivation makes people vulnerable to influence by others and impairs their ability to make complex decisions.*	77.2	85.5	75.9	79.7	16.06*
8	Explicit promises of leniency or immunity during interrogation can lead an innocent person to confess to a crime he or she did not commit.*	69.3	69.7	63.0	68.0	0.24
9	If police offer sympathy and morally justify a crime, suspects may believe that they will be treated with leniency upon confession.*	78.2	64.5	64.8	70.6	2.76
10	In decision-making, people are influenced more by immediate outcomes than by longer term consequences.*	82.2	82.9	75.9	81.0	4.79
11	Hearing incorrect information about an observed event can change people’s memory for that event.*	70.3	82.9	66.7	73.6	9.42
12	Individuals who have intellectual disabilities are particularly vulnerable to the pressures of social influence.*	81.2	60.5	51.9	67.5	7.81
13	People tend to underestimate how much other people’s behaviour is influenced by the situation they are in.*	80.2	84.2	74.1	80.1	7.37
14	Normal adults (without special vulnerabilities or mental impairments) cannot be induced to confess to crimes they did not commit.	38.6	64.5	61.1	52.4	20.56*
15	Videotaped confessions that show only the suspect look more voluntary than those showing both the suspect and interviewer.*	65.3	42.1	48.1	53.7	7.56
16	Individuals with compliant or suggestible personalities are particularly vulnerable to influence during an interrogation.*	74.3	77.6	59.3	71.9	7.63
17	Compared to adults, adolescents who are interrogated are at greater risk to confess to a crime they did not commit.*	69.3	59.2	48.1	61.0	7.00
18	People can be trained to know at high levels of accuracy when someone is telling the truth or lying by looking for nonverbal signs of anxiety.	11.9	25.0	24.1	19.0	1.11
19	Authority figures can get people to do bad things by first getting them to do something small, and gradually asking them to do worse and worse.*	76.2	84.2	75.9	78.8	11.52
20	People can be trained to know at high levels of accuracy when someone is lying or telling the truth by looking for fleeting and subtle ‘micro-expressions’ in the face.	14.9	17.1	27.8	18.6	2.33
21	Individuals with diagnosed psychological disorders are particularly vulnerable to influence during an interrogation.*	79.2	71.1	57.4	71.4	11.91
22	Threats of physical violence and punishment during interrogation can lead an innocent person to confess to a crime he or she did not commit.*	81.2	77.6	74.1	78.4	2.87
23	Conducting an interview in a public place, as opposed to interrogation room, will increase the risk that an innocent suspect would confess to a crime he or she did not commit.	37.6	44.7	40.7	40.7	2.76
24	Confessions can be verified as true by the details that they contain about the crime.	19.8	32.9	27.8	26.0	4.98
25	Innocent people never give false confessions ‘voluntarily’ – without pressure from police.	29.7	46.1	44.4	38.5	6.81
26	The rights to silence and a lawyer provide people with adequate protection from coercive interrogations.	12.9	23.7	29.6	20.3	1.92
27	The notion that someone can be induced to confess to a serious crime he or she did not commit is well known to laypeople as a matter of common sense.	21.8	38.2	38.9	31.2	9.70
28	Enhanced interrogation techniques (such as torture) can lead an innocent person to confess or otherwise provide unreliable information.*	80.2	90.8	72.2	81.8	15.06*
29	Interviewing techniques aimed at getting information from suspects without accusations lead to more true confessions and fewer false confessions compared to techniques that are more confrontational.*	73.3	69.7	57.4	68.4	1.11
30	Even without other evidence, people can accurately tell the difference between true and false confessions.	44.6	59.2	53.7	51.5	4.12

Note: Items marked with asterisks indicate that they are in line with current scientific consensus. *p < .002 (corrected for multiple testing).

Similar to the analyses for beliefs about memory, Kruskal–Wallis H tests were conducted to determine whether there were differences in endorsement scores for each statement among the three groups of participants. P-values have been adjusted to .002 using Bonferroni corrections; this was done by dividing the alpha level of .05 by 30 (i.e. the number of tests). As reported in Table 3, the three groups had significantly different responses on 5 out of 30 statements. According to post hoc analyses, psychology officers of the Ministry of Home Affairs were less likely to endorse statements that reflect current literature. Out of the 5 statements where there were differences across groups, the psychology officer group was the least likely to endorse statements that reflect current literature in four occasions (i.e. Statements 1, 7, 14 and 28), whereas the lay participant group was the least likely to do so in one occasion (i.e. Statements 2).

Beliefs about deception detection

The responses participants provided to the open-ended question about what they thought were good cues for detecting lies were first assigned into four main categories, namely speech characteristics, speech content, facial behaviours and body movements. This coding was done following Bogaard et al. (2016), in which the written responses were coded with reference to the categories employed by Akehurst et al. (1996), but for this study, the paper by Vrij et al. (2006) was also referenced due its relevance. Two research assistants were trained by the first author to organise the cues into the relevant categories. Each research assistant performed the coding independently; there was a good inter-rater agreement, Cohen’s κ = .78. (95% CI, .71 to .84). Differences of opinion in categorisation were resolved through discussions and consultation with the authors until consensus was reached.

Data were available for 82 psychology officers, 67 participants with a psychology background and 40 lay participants with no psychology background. Across three groups, there was a total of 395 reported cues believed to be indicative of deception. Out of the reported cues, 40.50% were body movements (e.g. self-manipulations, nervous body), 29.11% were facial behaviours (e.g. eye contact, variations in facial expression), 16.71% were speech content (e.g. contradictions, details) and 10.89% were speech characteristics (e.g. tone, stutter), while 2.78% did not fit into any of these categories (e.g. attention, unusual bank activity which indicates they are paid to lie, overall manner such as being defensive). A large proportion of the reported cues were not further specified, for instance, participants only listed broad cue categories such as ‘body language’, ‘facial expressions’, ‘verbal’ and ‘nonverbal’ cues. In instances where ‘verbal cues’ were reported, we categorised them into the ‘speech content’ category, whereas when ‘nonverbal cues’ were reported, we categorised them in the ‘body language’ category. Table 4 shows the frequency in which all these cues were reported. A chi-square test of independence was conducted between participant groups and type of cues reported. There was no significant association between participant groups and type of cues reported, χ²(8) = 7.68, p = .47.

Table 4. Frequency of cues being reported as indicators of deception by psychology officers, participants with psychology background and lay participants without psychology background.

Cues	Psychology officers (n = 82)	Respondents with psychology background (n = 67)	Lay people (n = 40)	Total (N = 157)
Body movements	64 (42.67%)	64 (37.87%)	32 (42.11%)	160
Facial behaviours	40 (26.67%)	51 (30.18%)	24 (31.58%)	115
Speech content	19 (12.67%)	33 (19.53%)	14 (18.42%)	66
Speech characteristics	21 (14.00%)	17 (10.06%)	5 (6.58%)	43
Others	6 (4.00%)	4 (2.37%)	1 (1.32%)	11
Total	150	169	76	395

Discussion

The current study examined the beliefs about memory, police interviewing and lie detection among people, and whether such beliefs were in line with the latest scientific literature. Comparisons were done across psychology officer serving the Ministry of Home Affairs, participants who have had or are undergoing training in psychology/ counselling/ cognitive neuroscience, and participants with no such background training.

Interestingly, while psychology officers reported as having engaged with memory literature more than participants without a psychology background, their beliefs around the topic were no more accurate. With regard to the investigative interviewing literature, there were no differences in overall ‘performance’ among the groups, even though psychology officers were more likely than the lay participants group to report reading relevant journal articles on investigative interviewing. It is worth noting that the mean ‘performance’ for participants who have had or are undergoing training in psychology or cognate areas seemed to be highest among all three groups. While it is known that the Ministry of Home Affairs psychology officers, who are civil servants, deal with a variety of clients including victims of crime and their families, prisoners and law enforcement officers, there was no information about the other participants’ line of work. Among the non-psychology officers, about half the sample consists of students currently pursuing psychology and cognate areas; this seems to imply that being engaged with (presumably) current psychological literature is what informs beliefs and knowledge about the topics. However, there was no further information about the background training that participants received, particularly which institutions they obtained their qualifications from (e.g. public vs. private, local vs. abroad, etc.), which would have been useful for examining this interpretation.

Analysis by item showed that the three groups of participants with different educational backgrounds performed differently on certain items. Again, there appears to be a pattern where participants who are or had been previously pursuing psychology and cognate areas at the time of the study held beliefs that are more in line with the state-of-the-art literature. There were several misconceptions held by most participants that are worth pointing out. In relation to the topic of memory, the controversial idea of unconscious – as opposed to conscious – repression of traumatic memories (Statement 1) is prevalent, but to date there is no reliable scientific evidence to support the occurrence of this phenomenon (see Otgaar et al., 2021). Qualities of memories, including its vividness (Statement 4), emotional content (Statement 10) and confidence (Statement 6) in reporting are not indicative of accuracy according to previous studies (e.g. Rimmele et al., 2012; Talarico & Rubin, 2003), but current findings show that many tended to endorse such beliefs. Today, memory scholars agree that memories are reconstructive as opposed to reproductive, and that memories are almost never exact replicas of events that have occurred (Howe & Knott, 2015), but in reality, a sizeable group of people still subscribe to the belief that memory works like a recorder (Statement 2). People also believe that we have memories from very early childhood (Statement 7), even though memories from age 2 or younger are likely fictional (Akhtar et al., 2018). The classic study by Loftus and Pickrell (1995) as well as more recent laboratory studies (see Muschalla & Schönborn, 2021 for a review) have found that people can indeed develop false memories that are indistinguishable from ‘real’ memories (Statement 9), but many are still not aware of this notion. The psychology officer group was the least likely among the three groups to think that memory can be influenced by suggestions (Statement 8) and that people can develop false memories even from non-traumatic events (Statement 3).

On the topic of police interviewing, it was generally believed that people (Statement 1), including the police (Statement 6), can become effective lie detectors through training in the use of nonverbal cues (Statements 18 and 20). However, the efficacy of such lie detection methods and training has yet to be established (Vrij & Granhag, 2012). In particular, psychology officers in the government sector were more likely than the other participants to believe trained police can accurately distinguish between truths and lies, even though existing studies (e.g. Akehurst et al., 1996; Delmas et al., 2019) have shown that police officers are generally no more accurate than ordinary civilians at lie detection. It was also generally assumed that one’s rights to remain silent and to a lawyer are adequate safeguards during interrogations (Statement 26), despite evidence (albeit from the United States) showing that certain populations such as young adolescents often do not comprehend their rights (e.g. Zelle et al., 2015). Of note, participants appeared to have misconceptions of statements surrounding the topic of confession. For instance, many were of the opinion that confessions can be verified through how detailed they are (Statement 24), even though to date there is no validated method to determine the veracity of a confession (Geven et al., 2020). Many also thought that conducting an interview in a public setting increases the risk of a false confession (Statement 23), but very few believed that false confessions are well known as a matter of common sense (Statement 27). Many practising psychology officers from the Ministry of Home Affairs also thought that normal adults cannot be induced to falsely confess (Statement 14). It is now known by psychology scholars that false confessions are not a new phenomenon, and they occur regularly in criminal justice settings all around the world (Kassin, 2012). Lay participants were the least likely among the three groups to be aware of false confession risk factors such as youth (Statement 2), sleep deprivation (Statement 7) and coercive interrogation techniques (Statement 28), suggesting the being engaged within the psychology field can increase awareness.

Taken together, it is evident from the findings that many practising psychology officers harbour misconceptions about how memory and false confessions work. This situation partially stems from the fact that the forensic psychology field in Malaysia is still very much in its infancy, with very few scholars in the country invested in the topic, evident through the lack of available research publications. This highlights the need for knowledge training in fundamental psychology topics such as cognitive psychology, as well as its applications.

There were no associations between the three groups and the types of cues they believed to be indicators of deception. According to Vrij et al. (2023), researchers have over the years shifted their attention to focussing on speech content to detect deceit, however many practitioners still tend to rely on nonverbal cues despite their unreliability as indicators of deceit. Findings from the current study appear to corroborate Vrij et al.’s observation, in that more than half of the reported cues across all participant groups were nonverbal in nature, and less than 20% were about verbal content. Again, psychological research in the area of lie detection as applied to police practice is not typically part of the mainstream psychology curricula in Malaysia – mainly due to the lack of expert knowledge in the field of forensic psychology – hence it would not be surprising of people to be generally unaware of its theoretical and empirical developments.

Some of the psychological concepts in question can be complex and abstract, and therefore cognitively taxing and time-consuming to absorb, even for expert academics who are often fully immersed in the literature. It is generally understood that the Malaysian civil service is often operating at maximum capacity, with very limited time for scholarship.

Having said that, regardless of their previous formal education experience, to ensure that these officers are armed with accurate knowledge and deliver evidence-informed psychological practice, investing in high-quality training, followed by effective evaluation, timely supervision and continuous education, remain a worthwhile endeavour. Equally important is the need for educators in the psychology field to ensure that the syllabi are carefully planned and regularly updated to be in line with international standards.

Limitations

Several limitations of the present study are noted. First, attention checks were not embedded so it was not possible to identify careless responding. Instead, participants that were responding extremely quickly (i.e. less than two minutes) were deemed to have answered the questions mindlessly and hence were excluded from the analyses.

Second, as mentioned in the Discussion, participants were not asked about which institutions they obtained their qualifications from (e.g. public vs. private, local vs. abroad, etc.). This is relevant because the cost of pursuing tertiary education in Malaysia varies significantly across institutions, which has a direct impact on the quality of education provided. The first two authors of this study are employed by a UK university with an international branch campus in Malaysia, while the research assistants were students from another UK university with an international branch campus in Malaysia; it is thus assumed that many of the participants from the non-psychology officer sample obtained their degrees from private institutions that must comply with UK programme standards. An inclusion of a comparison between public and private higher education providers would have been useful to understand whether there is a difference in how institutions engage with current scientific literature.

Further, participants were asked to self-report whether they ‘rarely’, ‘occasionally’ or ‘regularly’ read the literature. As operational definitions of these words were not provided to participants, it is difficult to determine how participants interpreted these words, and whether the interpretations were the same across participants.

Conclusion

More than a decade ago, Lilienfeld et al. (2010) argued that it is crucial for psychology instructors to routinely address misconceptions and encourage critical thinking during courses at the introductory level, so that students are well-equipped with skills to distinguish science from pseudoscience. This recommendation remains relevant today; the present findings and the existing literature suggest that extensive and well-designed teaching and training are needed to help dispel incorrect claims about human behaviour. These are especially pertinent to ensure rational decision-making in the justice system.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at and .

Acknowledgements

Special thanks to the Ministry of Home Affairs Malaysia for their support towards this project. Thanks to research assistants Vanessa Hui En Chua and Yan Zhen Tan for their assistance in data collection, cleaning, and coding.

Disclosure statement

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

Data availability statement

The data that support the findings of this study are available via https://osf.io/36w9y.

Notes

1 The mean response time among our respondents was 14.86 min (Mdn = 10 min). Participants that were responding extremely quickly (i.e. in two minutes or less) were deemed to have responded mindlessly.

2 The non-Malaysians were students who took part in the study through the University of Reading School of Psychology and Clinical Language Sciences participant pool management system, which recruits participants from both the UK and Malaysia campuses.