The role of video background cues in the virtual court: a psychological perspective

ABSTRACT

Remote appearances for courtroom proceedings have become common practice in recent years. When a court participant appears remotely, they introduce new and often tangential cues as part of their video background. We have seen varying (and at times, controversial) background cues across virtual court members, with scholars and legal professionals noting the potential effects these cues may have on decisions and impressions (e.g. Bandes & Feigenson, 2020). Advice is often provided for how people should appear to virtual court, but this is often made with no direct link to empirical evidence. In this piece, we review psychological literature exploring how and why background cues may influence impressions and decision-making. In particular, we examine the role of cognitive factors such as schema activation and consequences for memory and truth assessment. We further consider social factors such as the connection between physical spaces and personality in shaping impression formation and decision-making. Finally, we highlight future research opportunities to target unanswered questions and ongoing innovations in virtual courts, in addition to considering policies and procedures that can be implemented to avoid background cue effects.

KEYWORDS:

• remote appearances
• virtual courts
• background cues
• impressions
• decision-making

Estimates suggest that between 2020 and 2021, in the UK alone, over 990,000 virtual court hearings (largely civil) were held remotely (Spinney, 2022). In these remote hearings, people joined digital audio-visual platforms instead of physical spaces to participate in legal decision-making. Driven in large part by necessity during the pandemic, over 160 countries have implemented some form of remote hearing in the courtroom context (Remote Courts Worldwide, 2022). When people appear for a virtual court via video, they often share not only their faces, but also their surroundings – furniture, plants, decorations (or lack thereof), wallpaper, pets, and even children (Fosslien & Duffy, 2020). In a physical courtroom, key trial actors share a consistent backdrop and context. In a virtual court, this consistency and shared space has not been possible or procedurally implemented. Since the rapid flip to fully virtual courts, we have seen wild examples of contextual cues, including lawyers joining trials from bed and appearing with cat filters. We have also seen less jarring shifts in context that have perhaps become more normalized, such as variations in office or working spaces that may unwittingly signal socioeconomic status, general organization, access to resources, and cues about home environments. For defendants, who may have less control of their environment, some have even had to appear with obvious custodial cues in the background that may act to dramatically influence how they are viewed by others (e.g. perceptions of dangerousness; McKay, 2018; Rossner et al., 2017; see Devine & Caughlin, 2014, for review on extralegal factors).

In this piece we raise the possibility that we should attend to and consider policy and procedures regarding backgrounds in virtual courts. We first consider psychological research that helps us to understand how backgrounds may shape cognitive and social psychological processing by addressing (1) how background cues may affect memory for, and perceptions of shared content, and then turn to (2) how background cues may influence one’s impressions of others. We then consider the additive impact cognitive load in remote communications might have on these effects and examine what policies and procedures could be put in place to mitigate the potential impact of contextual cues in virtual courts. We also suggest key targets for future research that may serve to inform ongoing innovations in virtual courts going forward.

Remote appearances in the virtual court: policy, practice and procedure?

The use of remote appearances in the criminal justice system is not new. Many courtrooms around the world had been using virtual courts to some degree prior to 2020 (see Krans et al., 2020). Expert or vulnerable witnesses routinely joined proceedings using Audio-Visual Link (AVL) and in some jurisdictions, certain cases were frequently conducted fully remotely (see Legg & Song, 2021). Nonetheless, policy, procedure, and empirical research supporting fully virtual courts was lacking when the world flipped to remote hearings as a response to a worldwide pandemic (Nir & Musial, 2022). One symptom of this lack of policy and procedure is the varied advice offered to remote participants both across and within jurisdictions. Lawyers have been encouraged to appear in decluttered offices (Thomson Reuters Institute, 2021), while clients have been advised to appear with a neutral and plain background (Keogh et al., 2020; Rossner, 2021). Observational research findings capture varied formats in which people appear, with some seemingly having no choice but to attend from their workplace or prison cell (Nir & Musial, 2022; Thornburg, 2021). In one observational study, six law students observed one month of Texas family court legal proceedings and documented witnesses and clients appearing from their homes, work offices, attorney offices, physical courtrooms, and from jails (Thornburg, 2021). Another study (also in the USA) had 44 legal students document observations of criminal virtual courtroom proceedings, and documented that defendants most often appeared either in remote booths or prison cells (Nir & Musial, 2022).

In some jurisdictions the courts themselves have provided formal advice to ‘appear with a clean, neutral background’ (County Court Victoria, 2021, p. 1) despite allowing people to join from devices that may have varied video and audio quality, and awkward angles, all of which may negatively impact communication and incidentally signal social status (see for instance, Bild et al., 2021). Indeed, in joining the Australian County Court of Victoria (2021), one could use tablets, phones, computers, or even audio only. From the many examples listed here and shared in mainstream social media, there is clear consensus that a variety of backgrounds have been presented. But what does the existing empirical research show in terms of the relative effect of these backgrounds?

Throughout this review we highlight tangential factors that may influence people’s judgment and decision-making in courtroom contexts. However, our central focus is on the impact of visual features in virtual courts and their capacity to shape evaluations made in this space. Indeed, people are inclined to believe the world is as they perceive it (naïve realism; Granot et al., 2018; Ross & Ward, 1996). Further, although visual information is still subject to biases, people tend to believe they hold more accurate visual representations compared to other sensory inputs (Jones et al., 2018). In the context of the courtroom, Feigenson (2010) argues that visual information presented through evidence is particularly influential and memorable for jurors. While remote testimony differs from visual evidence (e.g. photographs from the crime scene; animation recreations), Feigenson (2010) argues for a broader focus on the impact of visual technology in the courtroom (see also the camera perspective bias, Lassiter et al., 2002).

While there are many features of a virtual court that depart from traditions, history, rituals, and symbols in a physical court space (Rossner, 2021), here we limit our analysis to background cues that have rarely entered the courtroom until now. In virtual courts, decision-makers, with the goal of being impartial, have rich cues about a person’s environment to overcome or ignore – among many other extralegal factors – when forming decisions. What impact might this have on a decision-maker’s social-cognitive processing of information? In the following three sections, we address: (a) how contextual cues may shape cognitive processes and (b) impression formation; (c) what is unique about virtual courts for a decision-maker; and (d) consider policy implications and questions that need to be answered moving forward.

How background cues affect memory for and perceptions of shared content

Broadly defined, background cues might be considered a form of additional contextual information that is not the primary target for judgment. In the case of virtual courts, the background is presented in concert with evidence a decision-maker is instructed to attend to. But the background itself is not probative of the credibility of the person, their evidence, or the extent to which it might be weighed. Thus, under these conditions the human brain is expected to process the evidence shared, without being influenced by the additional contextual cues behind the person providing the evidence. Research in cognitive and social psychology shows the converse: People are often influenced by additional contextual information that is not the primary target for judgment (e.g. Bild et al., 2021; Derksen et al., 2020; Sanson et al., 2020; Silva et al., 2019). In this section we describe at least three key mechanisms through which background cues may influence people's judgments.

Schema activation

When additional contextual cues are present during encoding or retrieval of an event or narrative, it can shape people’s memory for, and more immediate judgments about, information people encounter (e.g. Bransford & Johnson, 1972; Garry et al., 2007; Henkel, 2012). In short, people tend to remember details in line with knowledge networks or schemas activated at encoding and retrieval of an event (Anderson & Pichert, 1978; Collins & Loftus, 1975; Gerrie et al., 2006; Roediger & McDermott, 1995). Schemas are organized structures that represent knowledge networks about a given category. When activated, these semantic networks can lead people to fill in the gaps from memory, inserting details that are related to content initially encoded, but were never presented (Anderson & Pichert, 1978; Pichert & Anderson, 1977; Roediger & McDermott, 1995). For example, the Deese – Roediger – McDermott (DRM) paradigm shows that after encoding a list of semantically related words (e.g., dream, bed, rest) people can come to remember semantically or schematically-related, but non-presented, items (sleep). In more ecological instantiations of schema paradigms, people have come to remember information consistent with schemas activated in news stories. For instance, several studies show that adding a photo to a narrative can lead people to remember information (both true and false) that is semantically consistent with the photo presented (Garry et al., 2007; Henkel, 2012). In one study, people were asked to read a newspaper article about a weather event then, after a delay, recall information from the text (Garry et al., 2007). For some people, the article had a photo paired with it that depicted the aftermath of a storm, for others the accompanying photo captured a ‘before’ scene (a village before the storm hit). Those who saw the aftermath photo remembered more extreme descriptions of the event – sentences from the text that were never present in the newspaper article they just read. Henkel (2012) found a similar pattern. When people were shown photos depicting a likely conclusion of a story (e.g. a broken vase alongside a story stating the vase was simply ‘dropped’), they were likely to falsely recall this inferred conclusion being stated in the story. To the extent that schemas provide a kind of cognitive scaffold through which jurors can organize incoming information, they may act much like narrative frames as proposed in the story model (Pennington & Hastie, 1986; 1988). Thus, if backgrounds prime a given schema or way of thinking, this additional contextual information (that does not directly bear on the evidence presented) may shape how people comprehend and remember key evidence in a virtual court.

Conceptual fluency

Additional context can also incidentally shape more immediate judgments about information. When ideas and concepts are presented in a way that makes them easy to imagine or bring to mind, people use that ease of processing as a metacognitive cue to inform their judgments (Alter & Oppenheimer, 2009; Schwarz, 2015). In general, when information is conceptually fluent to process – easy to understand, imagine, and picture – people are inclined to find ideas and claims more plausible, likely to occur, and true (e.g. Garry et al., 1996; Sherman et al., 1985; see Schwarz & Newman, 2017). Several studies have investigated the role of photos as a form of contextual information that may shape conceptual fluency and subsequent judgment (for a review see Newman & Zhang, 2021). For example, photos that are only tangentially related to a claim can lead people to be more confident in the veracity of a claim, despite the fact they are noninformative of truth, an effect described as truthiness (Newman et al., 2012; see also Fenn et al., 2013). In a recent study on eyewitness statements, when participants saw a claim like ‘the man stole her wallet when she got off the bus’, along with a tangentially related stock photo of a bus, people were more inclined to believe the witness statement was true (Derksen et al., 2020). The photo did not provide any evidence of the claim, it simply related to the content and perhaps helped participants to more easily picture the event (see for a review Newman & Zhang, 2021). If contextual cues in the background facilitate the understanding or imagery regarding a witness account or a victim’s story, this may systematically bias judgment in favor of the description. Of course, the opposite may occur if the background is inconsistent or feels incongruent with the content being shared.

Metaphorical frames

The presence of additional contextual information in the form of background cues might also influence judgment by activating metaphorical frames. Knowledge networks and knowledge structures consist of not only literal associations (such as the color red being connected to the fruit apple), but metaphorical associations too (e.g. red also being connected to the emotion of anger; Allbritton et al., 1995; Lee & Schwarz, 2014). Indeed, conceptual metaphor theory suggests that people use metaphors to help guide decisions and judgments (Lakoff & Johnson, 1980). Conceptual metaphors can guide comprehension of new information and have consequences for how it is evaluated (Allbritton et al., 1995; Zhang et al., 2022). Contextual background cues might incidentally activate metaphors that have consequences for how information is processed. For instance, cues in environments that symbolize power (e.g. where a target is placed ‘up’ or demands an upward gaze) can lead people to assume higher authority (Schubert, 2005). Physical cues can prompt a similar process. When someone sits physically closer, people rate that person as being socially closer to them (Winter et al., 2018). Thus, incidental contextual cues in remote appearances – the color of a background, the distance and perhaps size of a person’s face on the screen, and position of a screen or camera – may all influence attributions about people through metaphorical associations. Backgrounds afford many more opportunities for metaphorical activation, such as a messy or tidy room (Muir et al., 2022), which may give rise to attributions about people’s personalities, a finding which we return to later in this review.

In a justice setting it is implicitly assumed that in a virtual court people are making decisions and attending to evidence presented, as in a typical court space. As noted earlier, this requires the brain to ignore/exclude other available context in cognitive processing and decision-making. As outlined above, there are several cognitive routes through which context might influence or bias people’s processing, memory for, and judgments of information. To what extent can humans overcome the impact of schema activation, conceptual fluency, or metaphorical frame in decision-making? Accumulating evidence suggests that we have a ‘bias blind-spot’ where we are often unaware of our biases, or at least fail to understand the magnitude of them, in how we form our judgments and decisions (Ehrlinger et al., 2005; Pronin, 2007; Pronin et al., 2004). If we are not aware of a bias, it can be difficult to correct. Even when we are warned or made aware of a bias, these warnings can have little effect: Sometimes ignoring the warning and claiming that we are objective in making decisions, especially compared to others (when we are not), or even engaging in overcorrection (Hansen et al., 2014; Pronin et al., 2002). Together this research suggests that contextual cues may have an insidious impact on how we evaluate evidence in virtual courts. In the next section we turn to social consequences and focus our analysis on how contextual cues can also shape our impressions of people who are sharing evidence.

How background cues influence impressions of others

The influence of contextual cues holds in settings more directly relevant for the virtual court, with an emphasis on evaluations of other people. The impression formation literature suggests that people form rapid judgments of others with very limited information. That is, humans tend to make snap judgments about other people within seconds of seeing a face (Willis & Todorov, 2006). These impression formations are often formed from ‘thin-slices’ of information which is often nondiagnostic to the judgment at hand (Ambady & Rosenthal, 1992), including (but not limited to) personal appearance (Naumann et al., 2009), facial details (e.g. babyfacedness, see Zebrowitz & Montepare, 2008), and personal style (Gillath et al., 2012). These thin-slices are also informed by contextual cues – cues that are not on the person, but signal where they are (e.g. Hwang et al., 2021; Maddux et al., 2005; Silva et al., 2019). For instance, seeing the same face across different contexts leads to different impressions about that person. In a study by Silva et al. (2019), participants saw photographs of faces that appeared either on a ‘hook-up’ website, ‘long-term dating’ website, or social website. Across three studies Silva et al. found that trustworthiness ratings were lower for (the very same) faces when they appeared on ‘hook-up’ sites, compared to the ‘long-term dating’ or social sites. These findings were based purely on the first impressions made of faces without any other identifying information (such as name or biography), showing that contextual cues (such as a background) can influence judgments regarding others.

Personal spaces and personality

More applied research shows that when contextual cues are not websites, but people’s own personal spaces, the effects are particularly apparent. Indeed, when we look at other people’s spaces, we make inferences about the person who uses that room (Gosling et al., 2002; Harris & Sachau, 2005; Johnson, 2017). Given that homes are free to be personalized by the owner, environmental psychologists suggest that traces of traits can be embedded through the way a home is decorated and kept (Harris & Brown, 1996). When a perceiver views a person’s home, they may take note of and use these personalization cues, shaping judgments regarding the owner’s personality (e.g. lens model, Brunswik, 1956). These cue judgments may be more prominent for first impressions – where there are no existing associations about the owner. Indeed, tangential cues tend to have more pronounced effects on judgment when people have less or ambiguous information about the target (Darley & Gross, 1983; Devine et al., 2009; Kahneman, 2011; Levinson & Young, 2009; Yi, 1990). Prior to remote communication, one would not usually enter a person’s home without knowing or having interacted with that person beforehand (to establish an invitation) – meaning they may have already formed some initial impressions of that person and their personality. Now, by connecting through video and showing backgrounds, one essentially ‘invites’ strangers into their home, as many forms of communications and first impressions are being made virtually.

To date, the experimental research that has explored these questions regarding one’s personal space and others’ impressions has focused on physical environments (not virtual). This research shows that people form directional impressions about a target person based purely on environmental cues in their living spaces. Gosling et al. (2002) showed that when strangers were shown other people’s personal workspaces (e.g. office) or living spaces (e.g. bedroom), they used environmental cues from the room in forming impressions of those people. The lens model describes these environmental cues (such as an organized desk) as a lens through which observers can link to judgments, such as personality traits (e.g. conscientiousness) of the room-owner (Brunswik, 1956; Gosling et al., 2002). In the Gosling et al. study, people with an organized physical space were rated as being higher on conscientiousness than those with less organized spaces. The presence of decorations led to people being rated as high in openness and extraversion. When Gosling and colleagues tested the owner’s personality profiles, they found that home environments may in fact reflect aspects of personality, and observers were actually more accurate in their impressions when shown a bedroom compared to a workplace office (where people may have less control of the environment). These findings are consistent across later studies (Fingerman et al., 2021; Harris & Sachau, 2005; Horgan et al., 2019; Johnson, 2017). Some of these studies involved observers walking into the physical room (e.g. Gosling et al., 2002; Horgan et al., 2019), and others simply showed observers photographs (e.g. Johnson, 2017) or descriptive text (e.g. Harris & Sachau, 2005) of the spaces, yet results remain consistent.

Much like the workplace office condition in the Gosling et al. (2002) study, in remote appearances, some people may have less control about how their background presents. This lack of control presents a specific kind of vulnerability for some. Research on remote learning, education, and the ‘digital divide’ highlights the possible negative and nondiagnostic impact of background for some speakers (Finders et al., 2021; García et al., 2020). With unequal access to technology, such as high-quality camera features and virtual background options, combined with unequal access to ‘quiet’ or ‘private’ places, speakers on the less resourced side of the digital divide may be evaluated negatively by their peers due to circumstance, rather than the quality of their contributions (Finders et al., 2021). On the flip side, people who have access to higher-quality technology and resources may receive more favorable evaluations (Ariturk et al., 2020). Research in mock juror paradigms supports these concerns. Indeed, a relevant meta-analysis highlights socioeconomic status as a well-established extralegal factor that jurors draw on when forming decisions (Devine & Caughlin, 2014).

Considered together, existing research in cognitive and social psychology suggests people do draw on background and contextual information when forming impressions and decisions. These impression formation effects, driven by contextual cues, might extend to virtual courts and lead to similar conclusions. Understanding possible contextual effects in a virtual environment adds several other factors that must be considered.

Factors unique to impression formation on remote appearances

As detailed above, research has shown that when people have the experience of walking into, or looking at a photo of, a room without any information about the resident, they make inferences. In a remote environment, however, the conditions of judgment change, as the target person is almost always present in the frame. While the research on people’s personal spaces and inferences made about personality shows robust effects (e.g. Gosling et al., 2002), this research has not, to our knowledge, been undertaken when other impactful cues about a person (facial features/clothing) are present. Will the same personality inferences produced through personal living and office spaces remain when the person is now in the picture? A further consideration is that in these studies, participants are explicitly instructed to form inferences about the residents based on their rooms. However, in virtual courts, decision-makers should only be attending to the evidence the person is presenting, not how their background appears, thus inferences connecting the two may be regarded as more implicit. These questions raise interesting avenues to explore in future research.

Further, in other studies on impression formation where people see both person and background information at the same time (e.g. Maddux et al., 2005), people are not explicitly informed that the background is one’s own personal space, signalling another area that warrants future research. Put simply, we do not know if seeing both person and personal space will produce similar (or additive) effects. This raises the question of how generalizable existing findings are to more remote appearance environments and highlights the need for further empirical research.

Another condition of judgment that changes in remote appearance contexts is that people are often under higher cognitive load (see Bailenson, 2021; Fauville et al., 2021). Indeed, there are concerns that being on remote platforms (e.g. Zoom) creates a load, leading to several challenges in a decision-making environment, but most relevant here, possibly increasing the tendency for people to rely on cognitive shortcuts in forming impressions of others (e.g. Bailenson, 2021; Jacoby et al., 1989; Tversky & Kahneman, 1989). When people process incoming information, they can rely on heuristic or systematic routes in decision-making (Kahneman, 2011; see also Chaiken, 1980; Petty & Cacioppo, 1986). Under more heuristic-reliant (also referred to as Type/System 1) conditions, people employ cognitive shortcuts to guide judgment, allowing cues such as feelings of familiarity and ease of processing to inform decisions. Under more systematic (Type/System 2) conditions, people engage in slower, more deliberate thinking, requiring more time and cognitive resources. With increasing cognitive demands, people may be less able to employ analytical assessments of content they encounter, relying instead on heuristics and biases to guide their judgments. Under these conditions, people may be more susceptible to variation in backgrounds and associated processing changes, or social-cognitive biases this extralegal factor may produce. These are research questions worth considering in future work.

There are several factors that may contribute to a higher cognitive load online than in person. When we communicate remotely, we are receiving information in an environment that is depleted of typical nonverbal information (e.g. appearing from the chest up can block vision of gestures) that would otherwise be present in such exchanges (Bailenson, 2021; Fauville et al., 2021). The presence of these cues typically facilitates communication, while the absence can create a cognitive burden, increasing working memory load (e.g. Goldin-Meadow et al., 2001). In online communication we also have to override an automatic tendency to look at people’s eyes, instead looking more directly at our own cameras to simulate eye contact (Bailenson, 2021; Bailenson et al., 2005; Kleinke, 1986). The lack of eye contact also raises additional concerns regarding inferences people may make about credibility and truth (see Bandes & Feigenson, 2020, 2021). The loss of typical in-person contextual cues can also influence our ability to interpret emotions (Aviezer et al., 2012a; see also 2012b). Aviezer and colleagues (2012a) find that, compared to when people only see a facial expression, additional cues regarding body positioning can enhance people’s judgments regarding whether someone is experiencing intense positive or negative emotions. We also have ongoing distractions on the screen to ignore, such as the self-view that appears on video conferencing software and warnings about internet quality (see literature on the effect of mirror/self-view, Fauville et al., 2021; Fejfar & Hoyle, 2000; Ingram et al., 1988). Together the depleted information environment and configuration demands – which require constant regulation of gaze and maintenance of attention – create a cognitively demanding communication exercise. Future research might consider whether potential cognitive load effects are moderated by the level of experience video conferencing users have with such technology.

Even without the additional resource demands created in virtual environments, operating as a decision-maker in a legal context can, alone, create cognitive complexity and limit working memory capacity, affecting judgments as a result (e.g. Payne et al., 2004). Trials can be inherently complex, requiring decision-makers to hold onto details while at the same time integrating new information, which can lead to increased reliance on more automatic associations such as stereotypes and heuristics to guide their decisions (see Kleider-Offutt et al., 2016, for review).

Considered more broadly, there are at least two key consequences to consider for the listener in virtual courts. First, the task of communication is cognitively demanding and may be magnified by the complexity of being a decision-maker in the first place. Second and relatedly, remote participants may be more likely to make inferences drawing on stereotypes and heuristics (e.g. Chaiken, 1980; Kahneman, 2011; Macrae et al., 1997; Petty & Cacioppo, 1986). This general finding of higher reliance on heuristics under cognitive load or with divided attention holds true in both the impression formation literature (e.g. Sherman & Frost, 2000), but also in literature that focuses on the evaluation of claims (e.g. Begg et al., 1992). The extent to which these patterns bear on the use of contextual cues is yet to be addressed in virtual environments but is another knowledge gap and worthy of further research.

Strategic use of backgrounds

Of course, in this paper we have focused on cases in virtual courts where backgrounds are not relevant to the evidence at hand, and how potential harmful biases may emerge (e.g. cues to SES and the digital divide; defendants in custody). However, through the same psychological processes, some backgrounds, if relevant to the evidence presented, may facilitate processing and understanding of the evidence. These processes could be used strategically. For instance, if a forensic scientist presented evidence with a backdrop photo that facilitates comprehension of their evidence, jurors may be able to better comprehend and remember that information later on. Such information may also feel more reliable, consistent with research that shows related (but nonprobative) photos can enhance truthiness of forensic science claims (Sanson et al., 2020).

Backgrounds may also be used for other strategies in the virtual court. Consider the potential strategies lawyers can employ for their defendants or witnesses when appearing before the court. Instead of being told to have a neutral background, those appearing in the virtual court may start to receive advice to put family photos in the background and emphasize a loving family, or cues to their expertise in a given professional context. While dogs and children in the background may initially seem like a distraction, consider the effect of this when decision-makers begin to form impressions regarding a defendant. As noted by Scigliano (2021), ‘If a judge could see a defendant’s children or meager living conditions before sentencing, she might better appreciate the consequences of her decision’ (para. 23). Lawyers already use strategies for in-person courtrooms, such as dressing the defendant nicely and even putting glasses on them (Merry, 2013). One can also see how these strategies may evolve in a virtual court where backgrounds may become a key detail in a defendant’s or witness’ appearance. Indeed, this more deliberate use of backgrounds may also impact civil and family court contexts.

Future directions and policy implications

Throughout, we have assumed across many accounts that people notice background cues, which thus may alter decision-making. But under some conditions, there is evidence that with a central goal in mind (e.g. counting passes between team players) people can miss otherwise rather salient features of a visual scene (e.g. a Gorilla walking behind those players; Simons & Chabris, 1999). Relatedly, how individuals attend to cues in remote appearances may also be a point for further consideration. Visual and selective attention, amongst inattentional blindness, can lead decision-makers to attend to different cues (see Granot et al., 2014; Kahan et al., 2012). Indeed, in trial contexts, a decision-maker’s identity, and whether the target person holds a similar or different identity, has been shown to impact how harshly the target person is evaluated (Granot et al., 2014). There is also evidence that identity can impact the likelihood of inattentional blindness occurring (Brown-Iannuzzi et al., 2014). These cognitive accounts are exciting areas for future research in virtual contexts more generally.¹

Leveraging technology

The presence of contextual cues in the form of backgrounds raises a potential concern for virtual court proceedings, given the possibility that such cues may shape impressions about a person and impact how evidence is attended to and remembered. While court participants can access advice on virtual court participation, guidance regarding backgrounds is not supported or directly linked to a body of empirical evidence. The common assumption is that a plain neutral background is preferred (e.g. Fosslien & Duffy, 2020). While this guidance at face value makes sense, a plain background also means ambiguity. A person sitting in front of a white plain wall does not provide any cues about whether they are appearing from their bedroom, kitchen, living room, or office. While this might not matter for work or educational settings, in a criminal justice setting, decision-makers may be inclined to search for location cues. If, for instance, a defendant appears in front of a white wall, decision-makers may not know whether they are appearing from their home or a custodial institution. Assumptions that they are appearing from custody may lead to negative attitudes, suggesting that ‘neutral’ backgrounds may not exactly correlate to ‘neutral’ impressions (see Rossner, 2021).

Alongside neutral backgrounds, another solution to avoid potential background bias is to provide people with blurred or virtual backgrounds. Many courts encourage and provide free court-themed backgrounds for attendees to use (e.g. County Court Victoria, 2021; Keogh et al., 2020), and blurring effects are available on most video conferencing software. These effects may allow viewers to focus more on the person, rather than the background, as in photography (MacDonald & Enns, 2012). However, unless someone has a green screen or neutral wall behind them, the design of these effects often come with flaws, such as faces or body parts disappearing amidst the background. Moreover, for many virtual background designs, it is also obvious that it is not a realistic setting (e.g. appearing on a beach) and can appear fake. Initial research shows that participants of a Zoom conference call (students in an online lecture) tend to prefer natural backgrounds over virtual backgrounds, with many perceiving virtual backgrounds as ‘unprofessional, frivolous, distracting, and false’ (Goethe et al., 2021, p. 276). These findings suggest that, although virtual backgrounds are beneficial in hiding details about one’s home, they are not necessarily met with positive impressions (see also Hwang et al., 2021). The description of virtual backgrounds as ‘false’ also suggests something more concerning – that a fake (or blurred) background may be related to feelings of distrust (see research on how distrust can lead to more skeptical cognitive processing; Kleiman et al., 2015; Schul et al., 2004, 2008). This distrust may be particularly important in a courtroom context, where how much one trusts an expert’s opinion, witnesses’ statements, defendant’s testimony, a judge’s expertise, or the overall fairness of the case, is relevant to the decision-making process. Empirical research is yet to be conducted to explore the impact of neutral, virtual/fake backgrounds with these psychology and law questions in mind.

Another aspect to consider is that although virtual backgrounds may help to control contextual information, noisy environments and possible drop-out or poor technology provide another kind of contextual cue not examined here in detail but which, nonetheless, can influence people's judgments (Bild et al., 2021). Children or dogs in the background can indicate a person is in their home, and loud noises, alarms, or the sounds of heavy doors opening and closing could indicate a person is in custody (e.g. McKay, 2018). With the digital divide in mind, poor technology may not be able to handle virtual backgrounds well, due to higher system resources and requirements (e.g. Zoom Support, 2022). While not the current focus of this review, further research is also needed in fully understanding how visual information is used in virtual courts. Other perceptual elements, such as how videos and cameras are set up (such as angle and distance) may also shape evaluations of remote testimony (Lassiter et al., 2002; see also Feigenson & Spiesel, 2011). There have been calls for courts to consider alternatives, such as the ability to access private video conferencing pods, reducing the impact for people who do not have access to optimal resources (see Tait & Tay, 2019). Such changes may pose significant resource demands, but offer digital equity.

Instructions

If the backgrounds themselves cannot, or should not (in the case of virtual backgrounds), be changed, then decision-makers could be warned of the potential biasing effects they provide. Instructional courtroom warnings have been studied for many years by researchers interested in their efficacy. On the one hand, the research shows that instructions may counteract against biases because jurors become aware of them and can be motivated to correct them (Elek et al., 2012; Wegener et al., 2000; Wegener & Petty, 1997). On the other hand, instructions to disregard biases are often ineffective and can even lead to overcorrection or a backfire effect (Lieberman & Arndt, 2000; Ruva et al., 2022; see also the bias blind-spot, as discussed earlier in this review; Hansen et al., 2014; Pronin et al., 2002, 2004). Overcorrection is where a juror overcompensates and shows a new bias in the opposite direction (e.g. a bias against the defendant leads to a bias favoring the defendant; Wegener & Petty, 1997; see also discounting literature, Oppenheimer, 2004; Oppenheimer & Monin, 2009). Instructions can also result in a backfire effect, which occurs when jurors rely more on a bias after being told to not use it (e.g. a bias against the defendant is used to form a stronger negative bias; Lieberman & Arndt, 2000; see also reactance theory; Wolf & Montgomery, 1977). Indeed, as noted earlier, the fact that background cues are visual may make them harder to correct than other biases (Jones et al., 2018), and research on instructions mitigating visual biases (e.g. the camera-perspective bias) suggests varying success (Elek et al., 2012; Lassiter et al., 2002). Empirical research is warranted to test the effectiveness of instructions regarding backgrounds before any policy guidelines are suggested.

The research considered thus far suggests that psychological science has barely scratched the surface in understanding the impact of virtual courts and background cues in human judgment. While concerns regarding background cues have already been raised by scholars (e.g. Bandes & Feigenson, 2020), most of these concerns have been based on theoretical arguments, with some limited observational findings (e.g. Nir & Musial, 2022). More specific observational data (e.g. through virtual courtroom archives) that explores background and contextual cues could provide more insight into some of the arguments raised here. Future research could also explore virtual court factors through experimental designs, manipulating different backgrounds such as the ones discussed in this piece (e.g. digital divide cues or personal living spaces; compared to proposed solutions such as neutral, virtual, or blurred backgrounds). With the limited research on virtual courts, it is currently challenging to offer evidence-based policy solutions. Thus, further investment in this broader research domain is warranted.

Summary

Increased use of virtual courts means that empirical research should focus on factors that may impact decisions made within these spaces, such as backgrounds. Psychological research suggests several mechanisms of how cues within backgrounds may affect memory for, and perceptions of, shared content, as well as how cues can influence impressions of other people. Backgrounds may unwittingly shape people’s memory and immediate judgments about information presented in the virtual space through mechanisms such as schema activation, conceptual fluency, and metaphorical framing. Backgrounds may also shape impressions of others, through cues suggestive of personality traits or even socioeconomic status, providing a vulnerability for those who cannot control their spaces. While research has found that personal living spaces can be linked to personality inferences (Gosling et al., 2002), and, separately, that contextual cues can impact impressions (Maddux et al., 2005), future research needs to explore whether combining these two factors produces similar or additive effects.

The unique communication environment for decision-makers simply attending a virtual court proceeding introduces other factors that bear on cognitive resources, such as depleted nonverbal information, eye gaze, and self-view, in addition to trials being inherently complex. These virtual conditions may make it more likely that participants draw upon heuristics, such as background cues, highlighting the importance of considering these factors in judgment and decision-making processes. As technology continues to evolve, there is also the added question of whether familiarity with video conferencing software moderates cognitive load effects or what type of virtual cues are used while forming impressions.

Backgrounds are an unavoidable factor in virtual communications that warrant further research. The lack of empirical evidence in this area highlights several important avenues for research (as well as avenues not explored here, see Hans, 2022, for juror selection issues), especially since remote appearances will likely remain in future courtroom proceedings. Although alternative backgrounds (e.g. neutral, blurred, or virtual) may be initial solutions to these problems, we suggest that these options may also be subject to biases, such as distrust. Caution should therefore be used regarding background and procedural consideration in virtual courts, at least until more empirical research has explored these unanswered questions.

Data availability

Data sharing not applicable – no new data generated.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We thank one of our reviewers for contributing this idea.