In his commentary, Verleger (1995) argues against the proposal that the P300 is due to phase locked theta activity (1995). He points out that most of the power of the P300 lies in the delta range. Furthermore, he emphasizes his opinion that without incorporating a variety of ERP-experiments, any theory about brain physiology and memory will be incomplete. In my reply I refer to data which indicate that regardless of the effects of other frequency bands, theta power significantly influences the amplitude and latency of the P300. With respect to incorporating results from ERP-experiments, I argue that what would be needed is a direct comparison between ERP-results and results from comprehensive spectral analyses. For reasons outlined below, a critical issue for such a comparison would be to determine frequency bands individually for each subject.
2. In a series of interesting experiments, Basar and his coworkers have documented that auditory and visual evoked potentials are composed of significant portions of theta and alpha power. Furthermore, these data provide evidence for the view that auditory and visual evoked potentials can be described in terms of a phase locked oscillatory response within the alpha and theta band (e.g., Basar & Schormann, 1995; Schormann & Basar, 1995; and the general review in Basar, 1992). The critical question, however, whether the typical P300 can be explained in a similar way, still remains an open question on the basis of these data. However, in a study focusing on the frequency decomposition of the P300, we (Klimesch, Schimke & Pienert, 1995) have found that although most of the P300 power lies in the delta range (as Verleger has pointed out correctly), latency- and amplitude- measures were significantly influenced by the theta band.
3. When considering the fact that theta may influence the P300 significantly although most of the P300 power is due to EEG delta activity, the crucial question is where theta ends and delta begins. I think most researchers will agree that the definition of the cut off point between theta and delta is quite arbitrary. In a variety of experiments we have found repeatedly that the alpha bands must be defined individually by adjusting the two alpha bands to mean individual alpha frequency which is used as the cut off point between the lower and upper band. In Klimesch, Schimke & Schwaiger (1994) we have found in addition to this result that only if theta too is adjusted individually to the alpha bands, can the reported results about theta and episodic memory (see the target article) be observed. This means that frequency bands vary individually and that for example, a subject with a very slow mean alpha frequency would also have a slow theta, which (by the traditional view) would already be in the delta range. This finding and the fact that the P300 becomes increasingly affected by slower frequencies going from the theta to the delta range are important reasons against the traditional (sometimes extreme) low pass filtering in ERP-studies.
4. Of course, Verleger is right in his opinion that the variety of interesting ERP-data he was summarizing are very important for our understanding of brain physiology and memory. However, incorporating these results into a theory of brain oscillations would require a completely new method of analyzing ERPs which nonetheless could easily be obtained. This new method would require us: a) not to use extreme low pass filtering; b) to define frequency windows individually; c) to compute filtered ERPs in a similar way Basar is doing but within individually determined frequency bands; and d) to compare the obtained results with traditional procedures. On the basis of this or a similar procedure, future research will provide valuable information about the usefulness of the proposed brain oscillation theory.
5. Finally it should also be taken into account that oscillatory processes might trigger changes in the overall threshold in large neural cell assemblies thereby generating slow potential shifts which are reflected by slow ERP components. This hypothesis is testable. What is required would be once again a systematic comparison between ERP-data and data from spectral analyses by focusing additionally on a comparison between pre- and post-stimulus processes.
Basar, E. (1992) Brain natural frequencies are causal factors for resonances and induced rhythms. In E. Basar & T. H. Bulock (eds.), Induced rhythms in the brain (p. 425-467). Boston: Birkhauser.
Basar, E., & Schormann, M. (1995) Functional aspects of evoked alpha and theta responses in humans and cats. Biological Cybernetics, in press.
Klimesch, W. (1995) Memory Processes Described as Brain Oscillations in the EEG-Alpha and Theta Bands. PSYCOLOQUY 6(6) memory-brain.1.klimesch.
Klimesch, W., Schimke, H. & Schwaiger, J. (1994) Episodic and semantic memory: An analysis in the EEG-theta and alpha band. Electroencephalography & Clinical Neuro-physiology, 91, 428-441.
Klimesch, W., Schimke, H. & Pienert, Ch. (1995) Frequency components of the P300. Manuscript in preparation.
Schormann, M. & Basar, E. (1995) Topography of alpha and theta oscillatory responses upon auditory and visual stimuli in humans. Biological Cybernetics, in press.
Verleger, R. (1995) Memory-related EEG Potentials: Slow Negativities, Priming Positivity, Recognition Positivity, and Dm. PSYCOLOQUY 6(27) memory-brain.3.verleger.