Bottom-up driven speechreading in a speechreading expert: the case of AA (JK023)

Ear Hear. 2005 Apr;26(2):214-24. doi: 10.1097/00003446-200504000-00008.


Objective: This case study tested the threshold hypothesis (Rönnberg et al., 1998), which states that superior speechreading skill is possible only if high-order cognitive functions, such as capacious verbal working memory, enable efficient strategies.

Design: In a case study, a speechreading expert (AA) was tested on a number of speechreading and cognitive tasks and compared with control groups (z scores). Sentence-based speechreading tests, a word-decoding test, and a phoneme identification task were used to assess speechreading skill at different analytical levels. The cognitive test battery used included tasks of working memory (e.g., reading span), inference-making, phonological processing (e.g., rhyme-judgment), and central-executive functions (verbal fluency, Stroop task).

Results: Contrary to previous cases of extreme speechreading skill, AA excels on both low-order (phoneme identification: z = +2.83) and high-order (sentence-based: z = +8.12 and word-decoding: z = +4.21) speechreading tasks. AA does not display superior verbal inference-making ability (sentence-completion task: z = -0.36). Neither does he possess a superior working memory (reading span: z = +0.80). However, AA outperforms the controls on two measures of executive retrieval functions, the semantic (z = +3.77) and phonological verbal fluency tasks (z = +3.55).

Conclusions: The performance profile is inconsistent with the threshold hypothesis. Extreme speechreading accuracy can be obtained in ways other than via well-developed high-order cognitive functions. It is suggested that AA's extreme speechreading skill, which capitalizes on low-order functions in combination with efficient central executive functions, is due to early onset of hearing impairment.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Attention
  • Discrimination, Psychological
  • Humans
  • Lipreading*
  • Male
  • Memory
  • Middle Aged
  • Phonetics
  • Semantics
  • Speech Perception*
  • Time Factors
  • Verbal Behavior