8.3.1 - Olds and Milner’s reinforcement study

The reinforcement question and design

The study asked whether a brain stimulus could act as a consequence that increased the behaviour producing it.

Positive reinforcement occurs when the addition of a consequence increases the preceding response. “Positive” means added, rather than pleasant in an everyday sense. Olds and Milner were interested in this reinforcing function, not simply movements elicited after stimulation. If a rat repeatedly operates a lever to obtain a stimulus, the stimulus may be maintaining its action.

The laboratory study used 15 male hooded rats, weighing approximately 250 grams at the start. Electrodes were surgically implanted at different brain locations under anaesthesia; the rats had three days to recover. The septal area, in the central forebrain beneath the corpus callosum and between the lateral ventricles, was one important location. Other placements included the cingulate cortex and mammillothalamic tract. A flexible lead allowed the animal to move in a Skinner box.

Pressing the lever completed a circuit that delivered electrical stimulation through the electrode. The key comparison was within each animal: acquisition, when pressing produced stimulation, versus extinction, when the same response produced none. Electrode placement varied between animals. Thus the procedure combined an experimentally changed response consequence with comparisons across stimulated sites; it was not a random sample survey or a drug-administration study.

Testing and location checks

The testing procedure connected an observable action to a verified brain location.

On the fourth day after surgery, an approximately one-hour pretest familiarised the rat with the box and stimulation. The experimenter sometimes placed it on the lever and adjusted the voltage to a noticeable behavioural threshold. These pretest data were excluded. Subsequent sessions typically contained about three hours of acquisition and half an hour of extinction. Total acquisition exposure differed between early and later animals, so the table compared the first six acquisition hours for every rat. No food or water deprivation was used to motivate responding, and electrical stimulation was the only experimental reward.

Responses were recorded automatically. The main table reported the percentage of time spent responding regularly, counting periods with at least one response every 30 seconds as responding. This is not the percentage of rats choosing stimulation, nor a direct pleasure rating. The extinction percentage provides a comparison because high lever activity alone might reflect a tendency to respond rather than an effect of the stimulus.

After testing, the animals were killed and their brains sectioned, stained and examined to identify electrode-tip locations. This histological check matters: an intended placement does not establish where a tip actually ended up. However, electrical stimulation can affect fibres and connected structures, so locating the tip does not prove that a single tiny region independently produces all reward.

Site-specific findings

The reinforcement effect depended on where stimulation occurred, rather than appearing uniformly throughout the brain.

The four septal placements produced acquisition percentages of 75, 92, 85 and 88, compared with extinction percentages of 18, 6, 21 and 13. The original report describes the septal results as the most dependable reward effect. Selected original findings illustrate the contrast: