Direct Comparison of Two Statistical Methods for Determination of Evoked-Potential Thresholds

Abstract

Several statistical procedures have been proposed as objective methods for determining evoked-potential thresholds. Data have been presented to support each of the methods, but there have not been direct comparisons using the same data. The goal of the present study was to evaluate correlation and variance ratio statistics using common data. A secondary goal was to evaluate the utility of a derived potential for determining thresholds. Chronic, bipolar electrodes were stereotaxically implanted in the inferior colliculi of six chinchillas. Evoked potentials were obtained at 0.25, 0.5, 1.0, 2.0, 4.0 and 8.0 kHz using 12-ms tone bursts and 12-ms tone bursts superimposed on 120-ms pedestal tones which were of the same frequency as the bursts, but lower in amplitude by 15 dB. Alternate responses were averaged in blocks of 200 to 4000 depending on the size of the response. Correlations were calculated for the pairs of averages. A response was deemed present if the correlation coefficient reached the 0.05 level of significance in 4000 or fewer averages. Threshold was defined as the mean of the level at which the correlation was significant and a level 5 dB below that at which it was not. Variance ratios were calculated as described by Elberling and Don (1984) using the same data. Averaged tone burst and tone burst-plus pedestal data were differenced and the resulting waveforms subjected to the same statistical analyses described above. All analyses yielded thresholds which were essentially the same as those obtained using behavioral methods. When the difference between stimulus durations is taken into account, however, evoked-potential methods produced lower thresholds than behavioral methods. Evoked potentials, Objective threshold determination, Chinchilla.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1994
Accession Number
ADA285182

Entities

People

  • James H. Patterson Jr.
  • Ted L. Langford

Organizations

  • United States Army Aeromedical Research Lab

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Air Force
  • Air Force Facilities
  • Army Aviation
  • Background Noise
  • Biological Sciences
  • Biomedical Research
  • Brain
  • Classification
  • Electrodes
  • Electrophysiological Phenomena
  • Frequency
  • Health Services
  • Information Science
  • Medical Personnel
  • Military Research
  • Statistical Analysis
  • Test And Evaluation

Fields of Study

  • Psychology

Readers

  • Auditory Neuroscience/Auditory Physiology.
  • Computational Modeling and Simulation