1030 PSYCHOLOGICAL ASPECTS OF MAN’S ENVIRONMENT
true. Thus, classical conditioning techniques can
be used to measure sensory thresholds in animals.A second major type of conditioning is known as oper-
ant conditioning. In operant conditioning, there is nothing
equivalent to the UCS; a response which occurs is fol-
lowed by some subsequent event, such as the occurrence of
food, which will modify the frequency of the response. The
desired response may have a low probability of occurrence
and procedures such as shaping and the creation of special
environments may be necessary. Shaping consists of rein-
forcing successively closer approximations to the desired
response. A common example of a special environment is
the Skinner box, a small enclosure whose most prominent
feature is the response mechanism, usually a bar or button.
The rat or pigeon in a Skinner box, pushing a bar or peck-
ing or at a lever to obtain food or avoid electric shock are
example of operant conditioning situations.
Operant conditioning procedures can be used for the
same purposes as classical conditioning methods, but they
have some advantages. Classical conditioning is relatively
difficult to establish, is easily disturbed by any change in
the environment and the CR ceases to occur (undergoes
extinction) rapidly once the UCS is removed. Operant con-
ditioning, on the contrary, is relatively easy to establish.
One reinforcement often makes a marked difference in
response rates. It is less sensitive to extraneous environ-
mental interference and it can be maintained for long peri-
ods of time with a very low reinforcement rate. Records of
cumulative responses to various reinforcement schedules
show definite regularities. Changes in these patterns pro-
vide a sensitive measure of the effect of various chemicals
on behavior and have been extensively used for this pur-
pose in drug research. Threshold measurements in operant
conditioning require a slightly more complex situation than
in classical conditioning if the substance to be studied does
not act as a reinforcer of behavior. However, by using it as
a cue to indicate when some positive reinforcement is to be
given, the thresholds of such substances can be studied by
operant techniques. Both operant and classical condition-
ing techniques are applicable to humans as well as other
organisms.
Other Simple Learning Situations Other types of learn-
ing situations which have been studied sufficiently to provide
a basis for the study of behavioral effects of environmen-
tal factors are maze learning, discrimination learning, and
a variety of simple verbal learning tasks. The former tasks
may be used with either humans or animals. The verbal tasks
are limited to humans. The complexity of the stimulus situ-
ation is generally greater in these tasks. Specifying the rele-
vant dimensions of even simple mazes for learning situations
is complicated, and an enormous and continuing effort has
been devoted to the study of the features of verbal stimuli
which affect simple learning tasks such as paired associate
learning (where people are presented with pairs of words and
then asked to recall one of the words when shown the other)
and serial learning (where subjects are asked to learn lists of
words). They are therefore less easy to quantify and compareacross experiments than are conditioning responses and thus
introduce additional problems for the study of environmen-
tal effects. However, with sufficient care, reliable measure-
ments can be obtained.
These tasks come closer than simpler tasks to being of
a complexity relevant to behavior in real life situations. It
is quite possible that some stimulus will have no measur-
able effect on a simple process, but will effect more complex
behaviors; the reverse is sometimes also the case; environ-
mental conditions which will affect simple and monotonous
tasks may not affect more intrinsically interesting tasks. For
this reason, learning tasks other than simple conditioning are
often studied when the effects of various environments are
under investigation.
More Complex Tasks The number of human perfor-
mances which have been studied by psychologists is very
large; no satisfactory classification has been worked out for
procedures going beyond the simplest. Various bases for cat-
egorization have been proposed; for example, sensory-motor
to verbal, depending on the type of response; or in terms of
complexity, from simple conditioning to complex problem
solving. Probably the best ultimate solution will consist of
a task analysis for more complex behaviors in terms of their
dependence on more simple processes, but this has not been
achieved in any general way at present. The particular type
of task employed and the dimensions along which it is to
be analyzed can be best considered at present in terms of
the particular experimental situation. If an investigator is
concerned with the effects of certain pollutants on complex
judgments required by some task, he will do well to con-
struct a task as much like the criterion task as possible and
test the effects by means of a suitable design. If his concern
is with performance in a vigilance situation, then a vigilance
type task should be selected. Generalizations from effects on
simple behaviors to more complex behaviors are hazardous
at present.
Standardized Tests of Behavior Standardized tests have
been developed for a wide variety of human abilities. IQ tests
are the most famous, but many others exist. Such tests have
as their primary purpose either selection in connection with
personnel work, or evaluation for counseling for guidance,
but they may be used in the same way as other tests for the
evaluation of the effect of environments on performance.
Tests of this sort may be divided into tests of general
ability, tests of specific ability, and tests of personality. Both
of the former have been used in connection with environ-
mental work. Occasionally, the latter have been also, though
their suitability for the purpose of evaluating temporary
shifts in emotional state is often dubious. Most standardized
tests are designed to measure some relatively permanent
characteristic. However, some of these characteristics, such
as various types of manual dexterity, are sensitive to various
environmental forces in a fairly straightforward way. The
effects of depressant drugs on personality tests are consid-
erably less clear. There is also a serious problem of valid-
ity with regard to some tests, especially personality tests.
The objective tests, such as the Guilford-Zimmerman or
the Minnesota Multiphasic Personality Index (MMPI), haveC016_012_r03.indd 1030C016_012_r03.indd 1030 11/18/2005 11:03:53 AM11/18/2005 11:03:53 AM