Logan: Basic Principles of Learning TRACE & TEMPORAL CONDITIONING If a momentary occurrence of a stimulus event precedes the occurrence of a US, conditioning will occur, and conditioning will also occur if a US simply occurs on a regular temporal schedule. In the standard laboratory classical conditioning paradigm, the antecedent stimulus not only precedes the occurrence of the US but it stays on for the interstimulus interval such that it is physically present at the time of the US/UR. But this is not necessary; the antecedent stimulus event may be of brief duration so that conditioning must presumably occur to some kind of sensory trace (memory) of the antecedent stimulus. This latter procedure with a brief S is called TRACE CONDITIONING. As far as experimental evidence goes, it actually makes little difference whether the antecedent stimulus continues throughout the interstimulus interval (called DELAYED CONDITIONING) or whether it is a momentary event . . . conditioning occurs equally well with both procedures. There is also a large amount of transfer between the procedures. That is to say, one can condition with a brief S and then test with a prolonged S, or condition with a prolonged S and then test with a brief S; in either case, a strong CR occurs to the test stimulus. These findings indicate that the effective event defining a stimulus is a CHANGE in the level of stimulation. We tend to notice when something happens but then quickly turn our attention to other things. Temporal conditioning has here been combined with trace conditioning because it can actually be viewed as a form of trace conditioning. In TEMPORAL CONDITIONING, as US such as food to a hungry dog is simply delivered on a regular time schedule such as every thirty seconds. After sufficient exposure to such a situation, the dog comes to salivate quite regularly just before the next food delivery. It is called temporal conditioning because it would appear that the dog is conditioned simply to the passage of time. A better analysis, however, is that each occurrence of the US not only serves to elicit the UR but also serves as a stimulus for trace conditioning. Obviously, clear temporal conditioning cannot occur if the intervals are irregular, and there are limits as to how long those intervals can be and still produce closely-timed conditioned responses. These limits can become quite long if the organism develops some regular, paced response to occupy the time. A dog, for example, might turn around, go sniff the corner of the box, and then lick a paw while waiting for the next food delivery. Humans typically count out the seconds. The fact that organisms can be conditioned to stimulus events that occurred some time in the past strongly suggests that we possess some kind of "internal clock". The most common way to conceptualize this is in terms of a STIMULUS MEMORY TRACE. It is assumed that a stimulus event does not exactly reproduce itself inside the organism: rather, some persisting activity is initiated that decays gradually over time. We can effectively tell how long it has been since the stimulus event actually occurred by how strong this decaying stimulus memory trace is at the moment. (It can be assumed that substantially the same kind of decaying memory trace process takes place even if the stimulus remains on, the memory trace being of stimulus onset.) Trace conditioning is a good context in which to emphasize the fact that a stimulus decrease is an effective stimulus event for conditioning. A brief stimulus not only come on, it also goes off, and conditioning could as well be conceptualized as being dependent on this offset as to the onset. This point is made stronger if the stimulus, rather than being very brief such that the onset and offset are almost simultaneous, lasts for a short time. Depending on the length of the stimulus, conditioning may be more evident to its termination than to its onset simply because the offset occurs in a better position with reference to the US. Specifically, if the antecedent stimulus started ten seconds before the US and went off about one second before the US, the offset is a better stimulus for timing the occurrence of the US. Many of the implications of this principle remain to be analyzed experimentally. Of course, humans frequently have external clocks (e.g., watches) to help us know what time it is and to pace our behavior around regularly-occurring events in our environment. Probably the most familiar instance of temporal conditioning is related to meal time, but equally important is our sleep-wakefulness cycle. Both going to sleep and waking up are conditionable responses that can become quite precisely tuned to our internal biological rhythms, but many people fail to take advantage of these freely-available possibilities. You can readily condition yourself to go to sleep very shortly after you go to bed at night and equally to get up very shortly after you go to bed at night and equally to get up very shortly after you wake up in the morning. TERMS: Classical conditioning, delayed conditioning, temporal conditioning, trace conditioning, interstimulus interval, antecedent stimulus, unconditioned stimulus, stimulus memory trace.