Discriminant analysis of the localization of aggression-inducing electrode placements in the hypothalamus of male rats

Publication Type:

Journal Article


Brain Res, Volume 260, Number 1, p.61-79 (1983)

DOI Name (links to online publication)



Aggression; Agonistic Behavior; analysis; Animals; AREA; ATTACK; behaviour; Brain Mapping; Computers; Electric Stimulation; Electrodes; Human; Hypothalamus; Male; MALE-RATS; methods; PART; physiology; Probability; RAT; Rats; RESPONSES; Support; Non-U.S.Gov


Over 400 sites in the hypothalami of 270 male CPB/WE-zob rats were electrically stimulated in order to induce fights between males. The localization of electrodes inducing fights seems to differ from the localization of electrodes in which no fights can be induced. The differences in localization were detected and tested by a non-parametric discriminant analysis. The results were plotted by computer in a stereotaxic atlas of the hypothalamus of the CPB/WE strain. The method delimits areas within the hypothalamus where the probability to induce aggression is high, intermediate or low. Moreover, the procedure allows discrimination between areas where the thresholds for attack behaviour are generally lower than elsewhere and where the fiercest forms of attack are induced. None of the areas delimited coincide with a classical subdivision of the hypothalamus. Parts of the perifornical, anterior, lateral and ventromedial hypothalamus seem to be involved. The methods developed here may help to relate stimulation-induced aggression to other characteristics of the 'aggressive' area which cannot be obtained directly from fighting rats such as cytological, endocrinological, biochemical or physiological data. In addition, the procedure may help to settle disputes on the specificity of the localization of neural substrates of other stimulation-induced behaviours. The methods to discriminate between overlapping 3-dimensional reconstructions validated here for aggressive responses, can also be applied to other types of stereotaxic data and other types of effects, such as electrical, hormonal or other physiological responses. They may be especially useful if the localization of the neural population involved is not yet known, and unknown current-spread or diffusion of substances complicates the interpretation of stereotaxic data