Random local temporal structure of category fluency responses

David J. Meyer, Jason Messer, Tanya Singh, Peter J. Thomas, Wojbor A. Woyczynski, Jeffrey Kaye, Alan J. Lerner

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


The Category Fluency Test (CFT) provides a sensitive measurement of cognitive capabilities in humans related to retrieval from semantic memory. In particular, it is widely used to assess progress of cognitive impairment in patients with dementia. Previous research shows that, in the first approximation, the intensity of tested individuals' responses within a standard 60-s test period decays exponentially with time, with faster decay rates for more cognitively impaired patients. Such decay rate can then be viewed as a global (macro) diagnostic parameter of each test. In the present paper we focus on the statistical properties of the properly de-trended time intervals between consecutive responses (inter-call times) in the Category Fluency Test. In a sense, those properties reflect the local (micro) structure of the response generation process. We find that a good approximation for the distribution of the de-trended inter-call times is provided by the Weibull Distribution, a probability distribution that appears naturally in this context as a distribution of a minimum of independent random quantities and is the standard tool in industrial reliability theory. This insight leads us to a new interpretation of the concept of "navigating a semantic space" via patient responses.

Original languageEnglish (US)
Pages (from-to)213-231
Number of pages19
JournalJournal of Computational Neuroscience
Issue number2
StatePublished - Apr 2012


  • Alzheimer's disease
  • Category fluency test
  • Cognitive impairment
  • Inter response times
  • Semantic memory
  • Statistical temporal structure
  • Weibull distribution

ASJC Scopus subject areas

  • Sensory Systems
  • Cognitive Neuroscience
  • Cellular and Molecular Neuroscience


Dive into the research topics of 'Random local temporal structure of category fluency responses'. Together they form a unique fingerprint.

Cite this