Is My Siri the Same as Your Siri? An Exploration of Users’ Mental Model of Virtual Personal Assistants, Implications for Trust

Tenhundfeld, Nathan L.; Barr, Hannah M.; O’Hear, Emily H.; Weger, Kristin

doi:10.1109/thms.2021.3107493

Cited by 16 publications

(16 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We conducted two experiments that provide new insights into how individuals form trust in multi-component systems. Previous findings noted that trust generalizes across systems (Geels-Blair, Rice, & Schwark, 2013;Mehta & Rice, 2016;Rice & Keller, 2010), but few previous studies attempted to evaluate users' mental models of systems to explain this generalization of trust (Tenhundfeld et al, 2022). We found that the magnitude of contagion effects can be partially explained by perceptions of similarity, relatedness, blame, and responsibility.…”

Section: Discussionmentioning

confidence: 50%

“…Examples of complex systems include multiple networked automated systems (i.e. systems of systems such as UAVs and virtual personal assistants), self-driving vehicles, and distributed teams which are reliant on multiple humans interacting with multiple automated components (Atchley et al, 2022, Reig, Forlizzi, & Steinfeld, 2019Tenhundfeld et al, 2022;Wallister, de Visser, & Shaw, 2016). In complex environments involving automation, a better understanding of how individuals target trust in systems is critical.…”

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

“…While it is true that things may be functionally related or similar, and yet not be viewed as part of a system, to determine whether SWT contagion effects can be attributed to users generalizing trust across a system, one must first define what a 'system' is exactly. This is a long-running problem in the literature, and one that has not yet fully been addressed (Doyle & Ford, 1998;Tenhundfeld et al, 2022). Objective definitions and frameworks exist which attempt to define 'system' (Rasmussen, 1997;Hall & Fagen, 2017).…”

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

“…Alternatively, if user mental models of systems are more expansive than would be warranted by an objective definition, we may expect to see contagion effects that extend well beyond what researchers expect would objectively constitute a system. It is, however, difficult to parse out what users' mental models may be, as they seem to vary on an individual basis when participants describe them (Tenhundfeld et al, 2022). As such, one way in which we can establish the user mental models of what comprises a system is to ask about the functional relatedness and similarity between components, rather than asking directly about specific relationships within a system.…”

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

See 3 more Smart Citations

System-Wide Trust, Graded System Distrust, or simple Anchoring/Priming Effects?

Atchley¹,

O’Hear²,

Barr³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Trust in automation is an important construct to consider when attempting to predict proper use of automated systems. Automation is being used in increasingly diverse and more complex ways. Previous findings report that trust can be generalized across systems, leading to contagion effects. However, it is unclear in the current literature how users form perceptions of the boundaries between systems. To address this problem, we conducted two experiments in which we considered how perceived similarity, perceived functional relatedness, attributed blame, and attributed responsibility influenced the magnitude of contagion effects for trust. Participants were presented with hypothetical scenarios in which an airplane environment is depicted. We found that participants tended not to have a contagion effect for clearly unrelated components (a farmer and microwave) while showing a contagion effect for those that are related, suggesting that previous findings including contagion effects may not be explained by an anchoring effect. The degree of difference in trust between the success and failure conditions was also predicted by similarity and relatedness of the components. Moving forward, more research should focus on understanding how users form mental models of system boundaries.

show abstract

Section: Discussionmentioning

confidence: 50%

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

Section: System-wide Trust Graded System Distrust or Simple Anchoring...mentioning

confidence: 99%

See 2 more Smart Citations

System-Wide Trust, Graded System Distrust, or simple Anchoring/Priming Effects?

Atchley¹,

O’Hear²,

Barr³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…User mental models reflected their beliefs about the application’s rules for choosing music, i.e., the models were mechanistic in nature. By contrast, virtual personal assistants (VPAs) such as Alexa and Siri express human-like personality traits that influence the user’s mental models accordingly and their engagement with the VPA’s recommendations ( Poushneh, 2021 ; Tenhundfeld et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

Trust in the Danger Zone: Individual Differences in Confidence in Robot Threat Assessments

et al. 2022

View full text Add to dashboard Cite

Effective human–robot teaming (HRT) increasingly requires humans to work with intelligent, autonomous machines. However, novel features of intelligent autonomous systems such as social agency and incomprehensibility may influence the human’s trust in the machine. The human operator’s mental model for machine functioning is critical for trust. People may consider an intelligent machine partner as either an advanced tool or as a human-like teammate. This article reports a study that explored the role of individual differences in the mental model in a simulated environment. Multiple dispositional factors that may influence the dominant mental model were assessed. These included the Robot Threat Assessment (RoTA), which measures the person’s propensity to apply tool and teammate models in security contexts. Participants (N = 118) were paired with an intelligent robot tasked with making threat assessments in an urban setting. A transparency manipulation was used to influence the dominant mental model. For half of the participants, threat assessment was described as physics-based (e.g., weapons sensed by sensors); the remainder received transparency information that described psychological cues (e.g., facial expression). We expected that the physics-based transparency messages would guide the participant toward treating the robot as an advanced machine (advanced tool mental model activation), while psychological messaging would encourage perceptions of the robot as acting like a human partner (teammate mental model). We also manipulated situational danger cues present in the simulated environment. Participants rated their trust in the robot’s decision as well as threat and anxiety, for each of 24 urban scenes. They also completed the RoTA and additional individual-difference measures. Findings showed that trust assessments reflected the degree of congruence between the robot’s decision and situational danger cues, consistent with participants acting as Bayesian decision makers. Several scales, including the RoTA, were more predictive of trust when the robot was making psychology-based decisions, implying that trust reflected individual differences in the mental model of the robot as a teammate. These findings suggest scope for designing training that uncovers and mitigates the individual’s biases toward intelligent machines.

show abstract