Modeling panel detection frequenties by queing system theory: An application in gas chromatography olfactrometry.

J.H.F. Bult, B. van Putten, H.N.J. Schifferstein, J.P. Roozen, A.G.J. Voragen, J.H.A. Kroeze

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)

Abstract

In continuous vigilance tasks, the number of coincident panel responses to stimuli provides an index of stimulus detectability. To determine whether this number is due to chance, panel noise levels have been approximated by the maximum coincidence level obtained in stimulus-free conditions. This study proposes an alternative method by which to assess noise levels, derived from queuing system theory (QST). Instead of critical coincidence levels, QST modeling estimates the duration of coinciding responses in the absence of stimuli. The proposed method has the advantage over previous approaches that it yields more reliable noise estimates and allows for statistical testing. The method was applied in an olfactory detection experiment using 16 panelists in stimulus-present and stimulus-free conditions. We propose that QST may be used as an alternative to signal detection theory for analyzing data from continuous vigilance tasks.
Original languageEnglish
Pages (from-to)1125-1146
JournalPerception and Psychophysics
Volume66
Issue number7
DOIs
Publication statusPublished - 2004

Keywords

  • rehydrated french beans
  • signal-detection-theory
  • sniffing port analysis
  • odor thresholds
  • vigilance decrement
  • bell peppers
  • flavor
  • identification
  • detectability
  • probability

Fingerprint Dive into the research topics of 'Modeling panel detection frequenties by queing system theory: An application in gas chromatography olfactrometry.'. Together they form a unique fingerprint.

Cite this