Abstract
The most frequently used measures in behavioral research are distribution-determined measures that provide a quantitative determination of the relative frequency of a specified behavior. These measures are, however, insensitive to the sequence of behaviors. Complexity measures address this deficiency. Several sequence-sensitive measures including the topological entropy, metric entropy, algorithmic complexity and stochastic model complexity will be described. Several applications to psychiatric research will also be discussed. These applications include the characterization of changes in animal behavior in response to CNS-active drugs and the analysis of single-unit interspike interval spike trains. Clinical applications include the analysis of electromyographic and electroencephalographic signals, the examination of choice task behavior in clinical populations, and protocol analysis.
| Original language | English |
|---|---|
| Pages (from-to) | 408-416 |
| Number of pages | 9 |
| Journal | Molecular Psychiatry |
| Volume | 1 |
| Issue number | 5 |
| Publication status | Published - 1996 |
Keywords
- Algorithmic complexity
- Choice task paradigm
- Electroencephalograms
- Electromyograms
- Interspike intervals
- Metric entropy
- Protocol analysis
- Stochastic model complexity
- Symbolic dynamics
- Topological entropy