Real-Time Conversation Comprehension in Younger and Older Adults: The Effects of Noise, Visual Speech, Speaker Switch, and Context.
Chris Davis, April Ching, Afrah Haroon and 1 others
Seeing the speaker helps both younger and older adults follow a conversation, and older adults actually benefit more from visual speech than younger ones.
Background noise slows everyone down. Both age groups use context to predict upcoming words equally well.
Key findings
1Older adults showed a larger visual speech response time benefit than younger adults (Age Group × Modality interaction, p = .002), with the age difference significant in audio-only quiet (93.18 ms, p = .042) but not in audio-visual quiet (p = .875).
2Both age groups showed a response time cost when the speaker changed (p < .001), but the interaction between switch cost and age was not significant (p = .743), so the data do not confirm that older adults are disproportionately affected by speaker switches.
3Both age groups showed similar response time slowing for less predictable words (younger: 13.5 ms per bit of surprisal, older: 15.3 ms per bit), with no significant difference between groups (p = .276).
to read the rest of this summary — three full summaries a month, no card.
Already have an account?
What it does not show
Only 26% of the older adult participants had pure-tone audiometry; two participants with self-reported hearing difficulties were included, so it is unclear how much of the age-related pattern reflects peripheral hearing rather than central processing. The older adult sample was highly educated (18 of 19 had tertiary qualifications) and drawn from a university research register, so they are not representative of the broader older population and may have been more motivated to perform well. The authors explicitly describe the study as exploratory and preliminary, noting that procedures need to be made more rigorous. The category monitoring task requires participants to actively scan for category members, which may encourage a more systematic listening strategy than natural conversation and may not fully reflect everyday listening. Sample sizes are small (23 younger, 19 older), limiting the power to detect moderate age differences, particularly in the switch-cost and surprisal analyses where interactions were not significant.
Declared interests
Funded by the Australian Research Council (grant DP 200102188). The authors declared no potential conflicts of interest.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →