Skip to main content

Table 3 Key changes to acceptance-inhibiting cues to promote perceptions of VHA acceptability

From: Key changes to improve social presence of a virtual health assistant promoting colorectal cancer screening informed by a technology acceptance model

Cue

Key themes

Key changes

Version

Clothing

Not professional (e.g., scrubs)

Updated the VHA so she was dressed in business casual clothing rather than scrubs

Print to V1

Real/Fake

Low quality, animated, robotic looking

Changed lighting to introduce depth in visual features

V.1 to V.2

Scary/Creepy

Looked like a vampire (e.g., "fangs")

Added shading to mouth and teeth

V.1 to V.2

Expertise

Preference for doctor vs. nurse or lay health worker

Added white coat (e.g., business casual was too casual)

V.1 to V.2

Authority

Preference for middle age (e.g., too young = not enough knowledge vs. too old = not enough current knowledge)

Removed gray hair

V.1 to V.2

Trustworthiness

Desire for accurate, relevant information, desire to not feel targeted

Removed color-coded response options (e.g. red = “no”) to prevent perception of judgment when answering questions

V.1 to V.2

Navigability

Perceptions of how easy it is to use and navigate through the app

Added pause button with ability to tap to pause. Updated text size of subtitles. Removed user transition from waiting room to clinic room

V.1 to V.2

Movement

Unnatural movements, excessive hand gestures and rocking

Used motion capture suits to update motion to correspond with script (e.g. breathing animation)

V.2 to V.3

Appearance

More feminine, more dignified

Changed hairstyle, added jewelry

V.2 to V.3

Friendliness/likability

Angry looking, stressed out, not approachable

Added smile, removed furrowed brow

V.2 to V.3

Interactivity

Poor eye contact, low interactivity, limited opportunity to ask questions or have responses tailored to personal needs

Focused eye gaze, added randomness in eye movements (e.g., static to dynamic), new response option for health behavior questions (e.g., “yes, occasionally”), reduced extra info in VHA script

V.2 to V.3

Voice

Reading from a script, too fast/ loud, persuasive intent, subtitles not synched with audio

Selected race and gender concordant voice, adjusted subtitle speed, hired professional voice actors to record script, presented options to users

All