Explainability's effect on trust in AI-enabled systems is mixed: it can increase trust and reliance but also raises cognitive load, slows performance, and sometimes does not improve trust¶

Claim · All claims
Evidence · 1 study (2 entries) · quant-synthesis r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Several reviewed studies found explainability positively increased user trust and reliance, particularly in complex visual tasks with heatmap-based explanations. → Abramson SL 2026 q2 i? Other reviewed studies found explanations, particularly multiple types simultaneously, increased cognitive load, slowed user performance, or did not improve trust even when present. → Abramson SL 2026 (2)

Evidence¶

Abramson SL 2026¶

Abramson SL, Sora-Cardenas J, Nandakumar P, Bruce JE, Gopinath N, Zangari LG and Feijóo-García PG (2026). Trust and interaction design in AI-enabled systems: a systematic literature review. Front. Psychol. 17:1932655. https://doi.org/10.3389/fpsyg.2026.1932655

q2 · i? · quant-synthesis · r2

The review reports, citing Maehigashi et al. (2024a, b), that heatmap-based explanations aided comprehension and increased trust in complex visual tasks; the review narrates these studies without printing effect sizes.

"Several studies found that explainability positively increased user trust and reliance, particularly in complex visual tasks where heatmap-based explanations aided user comprehension of AI reasoning"

Abramson SL 2026 (2)¶

Abramson SL, Sora-Cardenas J, Nandakumar P, Bruce JE, Gopinath N, Zangari LG and Feijóo-García PG (2026). Trust and interaction design in AI-enabled systems: a systematic literature review. Front. Psychol. 17:1932655. https://doi.org/10.3389/fpsyg.2026.1932655

q2 · i? · review · r3

The review attributes countervailing findings to the same cited studies: explanations "increased cognitive load", slowed performance, and in some cases did not improve trust. No effect sizes are printed for these outcomes.

"other studies reported that providing explanations, particularly multiple types of explanations simultaneously, increased cognitive load ( Maehigashietal., 2024a ), slowed user performance (Maehigashietal.,2024b ),andinsomecasesdidnotimprovetrust evenwhenexplanationswerepresent"

Discussion¶

Learner Variables¶

• Processing Speed — outcome: instruction changes it

Related Claims¶

• Manipulating anthropomorphism shows no clear causal effect on trust in controlled post-generative-AI studies, though long-term use and cue type condition its effects — related
• Nine interaction design factors recur in post-2023 research on trust in AI-enabled systems, with explainability the most studied — related
• Error timing and stakes shape trust in AI systems: later errors may be more damaging, early errors may matter more, and trust recovers quickly after good post-error performance, with high-stakes errors causing abrupt collapse — related
• Text-based conversational interfaces were preferred over speech or embodied formats in healthcare AI trust studies, attributed to perceptions of reliability — related
• Verification bottleneck: as reliance on AI increases, accuracy in detecting AI errors decreases while confidence does not — related
• Students used visualizations to verify their own reasoning rather than as primary explanation tools, valuing control over their cognitive effort (agency) — related
• Instructors frequently encounter suspected inappropriate AI use but lack confidence in their current investigation processes — related
• Review-attributed evidence: user confidence in model answers exceeds model internal confidence, and explanation design narrows the gap — related
• Boundary conditions on verification and reliance include prior knowledge, learner characteristics, task stakes, verifiability, costs, accountability, AI system configuration, and multidimensional AI literacy — related
• User acceptance of AI-driven decisions depends on explanation type, with sensitivity-analysis and certification explanations increasing acceptance — related
• Supporting user interaction and control in the indicator implementation process positively affects transparency, trust, satisfaction, and acceptance in LA — related