4. Good-enough processing
People sometimes misinterpret the sentences that they hear or read. One possible reason for that is a race between slow algorithmic processing and “fast and frugal” heuristic good-enough processing that serves to support fast communication but sometimes results in incorrect representations. Heuristic processing can be both semantic, relying on semantic relations between words, and structural, relying on structural economy.
We aim to investigate how people’s age and linguistic noise influence the reliance on the good-enough language processing strategy during reading and during auditory speech perception. Overall, we tested 849 Russian-speakers (259 adolescents at the age of 13-17; 425 young adults at the age of 20-40; 165 older adults at the age of 55+). The participants either performed a self-paced reading task with comprehension questions, or listened to sentences and answered comprehension questions, or participated in an eye-tracking experiment (only young adults).
We found that all participants relied on good-enough processing strategy both during reading and auditory speech perception. Importantly, we showed that good-enough processing was faster than algorithmic processing. In reading, we found that the use of heuristics increased across the lifespan: adolescents did not rely on structural heuristics, in contrast to young and older adults; at the same time, older adults relied on semantic heuristics more than young adults and adolescents. In speech perception, the reliance on good-enough processing was found to be non-linear: young adults relied on semantic heuristics more than adolescents and older adults; at the same time, adolescents relied on structural heuristics more than young and older adults. Finally, our results showed no evidence that the participants relied on good-enough processing more in noisy conditions compared to no-noise conditions. That does not support the prediction of the noisy-channel approach to sentence comprehension.
The project was supported by RFBR grant №18-012-00640 (2018 - 2020), principal investigator — Anastasiya Lopukhina.
https://osf.io/6dfxh (oral speech perception)