DOCUMENT RESUME

ED 039 752 56 EM 008 088

AUTHOR Gagne, Robert M.

TITLE Learning Theory, Educational Media, and
Individualized Instruction.
INSTITUTION Academy for Educational Development, Inc.,
Washington, D.C.
SPONS AGENCY Office of Education (DHEW), Washington, D.C. Bureau
of Research.
BUREAU NO BR-8-0571
PUB DATE [701
NOTE 22p.; This is one of the support papers for "To
Improve Learning; a Report to the President and the
Congress of the United States by the Commission on
Instructional Technology", ED 034 905

EDRS PRICE EDRq Price MF-$0.25 HC-$1.20

DESCRIPTORS *Individualized Instruction, *Instructional Media,
*Learning Processes, *Learning Theories

ABSTRACT
Instruction and learning encompass more processes
than are included in learning theories themselves. Instruction
involves gaining and controlling attention, stimulating recall,
guiding the learning, providing feedback, arranging for remembering,,
and assessing outcomes. These functions are performed by various
media of instruction, but ultimately by the learner himself. Learning
is, after all, an individual matter. It is unlikely that one single
medium is best fitted to perform all the functions of learning. It
seems likely that carefully designed combinations of media may be
required to achieve the kind of instruction that is now effective,
and which at the same time exploits the properties of media to the
best advantage. (Author/GO)
 U.S. DEPARTMENT OF HEALTH, EDUCATION & WELFARE
OFFICE OF EDUCATION

THIS DOCUMENT HAS BEEN REPRODUCED EXACTLY AS RECEIVED FROM THE

POINTS OF VIEW OR OPINIONS
PERSON OR ORGANIZATION ORIGINATING IT,
STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATION

POSITION OR POLICY.
1 r

Learning Theory, Wucational Media,

CIO r.
Lir\

Cr4
and Individualized instruction
,Robert M. Gagnb*

0:=)

tO
LU

NO
Instruction of college and university students is an activity not
.customarily derived in a deliberate fashion from theories about learning.
Most college instructors set about their initial task of teaching courses
by using a model derived from their own college experiences; in other
words, they try to emulate their own professors. The new instructor may
spend many hours in selecting a text and other references, in planning
what he will say to his class of students, in seeing how certain topics
will "fit" a semester of so many weeks. But the question of just what
the students are going to be doing during these weeks, and how their
a Amities are going to affect their capabilities, is not likely to be
given a great deal of thought.
In proceeding in this manner to face the task of college instruction,
it is obvious that the new instructor is perpetuating many traditions. He
is planning his work in terms of the content of knowledge to which students
will be exposed, the kinds of communication he will make to them in lectures.
He is selecting for students a minimal set of readings and oral communica-
tions to which they will be "exposed." He is thinking in terms of how much
reading material and orally-presented material his students may bt' expected
to "absorb" during a given period of weeks. All of these activities are
traditional in the sense that they are the same ones he himself was subjected
to; they resulted in the framework for instruction as he experienced it.
It is also true that this traditional system may be said to "work."
The young instructor knows that, because it has worked for him and for most
of his fellow students. Why does it work? Under what circumstances does
it work? One suspects that it works within the confines of two major con-
O'dons: (1) first, that students attending college are highly selected to
accomplish learning in just this fashion; and (2) second, that what they
are expected to accomplish represents a limited set of educational goals.
u ..VM, .1 OA

* Robert M. Cagndis professor, Department of Educational Research,

College of Education,
Florida State University. This paper was prepared for faculty seminars sponsored
by B,Iro.mell Un4versity and the U. S. Office of Education on ,using educational
 To treat these questions fully, and to deal with all of their implica-
-t. Adns would require a different direction than the one this paper is-4

supposed to take. I shall therefore have to be content to suggest the lines

of questioning that seem to me to be opened up by identifying these limiting
conditions of traditional college instruction. Do we want to select just
those students for college who are most able to learn by traditional means?
By our selection procedures, are we simply perpetuating patterns of thought
and learning that are first laid down in high school and earlier? Are we
in danger of screening out by such procedures many individuals whose poten-
tial contributions to our culture are the most unusual? Are traditional
'Methods 'of instruction best adapted to prepare the student for the activi-
tieb of graduate school, where greater independence of thought is expected?
Are these methods, in fact, preparing the student to be both an independent'
thinker and a continuing learner?
If one admits these kinds of questions, into his thinking about the
nature cf college instruction, one faces the problem of understanding the
nature of instruction itself, and in particular, what instruction has to
with human learning. One is led to examine the ways in which things,
events, and ideas about them are presented to the human learner; in fact,
the ways in which relevant stimulation impinges upon the learner from his
to
environment. Further, one' is led to a consideration of what happens
.this stimulation when it reaches the nervous system of the learner - - in
oth6r words, what kinds of transformations it undergoes. For we know that
this environmental stimulation is processed in at least several different
ways: this is the kind of inference we make when we say that the human
individual has changed in the sense that he has learned something.
Media. The first,of these problems of stimulating the human learner,
regoesents the area of media of communication. Generally, we tend to des-
cribe media in terms of the material things that provide the vehicles for
the "messages" - as,, textbooks, newspapers, .blackboards, motion picture
projectors, television systems. However, for the purposes of considering
their effects on learning, there are advantages to attending instead to
the kinds of channels they offer. Considered in this way, one may conven-
iently describe media in several major categories as follows:
(1) actual objects and events
(2) veridical pictures (static and moving)
(3) diagrammatic pictures
.(4) printed language
(5) auditory language.
 _3_

These are the different ways in which the learner is affected by media.
De may be stimulated by actual objects and events, and a reasonable portion
of his learning results from such stimulation. Once he has learned how, in
his early years, the learner may be stimulated with apparently equal effect
by pictures, whether he sees them in a textbook, on a movie or television
screen. Again, following some early learning, he responds to diagrammatic
pictures, which are of several varieties. He responds to a two-dimensional
representation of a cube as if it were a cube, for example; and in a more
abstract way, he comes to understand the communication of a bar chart or
line graph. As.schooling proceeds, learning comes to depend increasingly
on the stimulation provided by printed language. There is surely much
truth in the definition of a university as a collection of books; even
though one recognizes this to be an ironically partial truth nowadays.
Auditory language has always been another major source of information for
use in learning, whether presented by itself as in a lecture, or combined
with the pictorial mode as in a motion picture, or. television program.
,
These are the ways, then, that stimulation is pre-
sented to the human learner. The second part of the problem to be considered
concerns what happens to this stimulation when it reaches the learner. How
is it transformed in such as way as, to change his capabilities from one
state to another? What kind of processing does it undergo in leading his
professors to conclude that he has learned?
Obviously, this is the area of learning theory. Psychologists have
.studied, experimented upon, speculated about, and generally tried to under-
stand learning for many years. Progress has not been rapid, but it surely
appears to have been speeded by the application of experimental methods in
use for about the last sixty years. As is.not unusual with phenomena of
living things, learning is a complicated process, occurring in many varie-
ties, forms, and situations. It is necessary first, therefore, to recog-
nize that learning theory as it exists today is alhighly inelegant and
unfinished entity. Nevertheless, there do appear to be some fairly funda-
mental and stable principles which serve to tell us what learning is not
like, and to suggest the outlines of what it is like.
Sorting out the general principles from the more specific ones in
learning theory is by no means an easy task. Similarly, selecting those
principles of ,learning which are most highly relevant to the practical
 .

pr6151em of instruction is not a self-evident procedure. The reason for

;his is that modern learning inVestigators have chosen different models
to study, and they are intent on accounting for these models. Sometimes,
these models resemble the learning of the school child, or the college
student, and sometimes they do not. For example, the prototype learning
;Situations represented in a recent influential book on Catepories of Human
Itallika, *(Welton, 1964), are approximately as follows
(1) Conditioning: Learning to blink the eye to a signal
(2) Rote learning: Learning to memorize pairs and lists of
Words
(3) Probability learning: Learning to choose a correct
alternative from a set of Words or objects
(4) Short-term memory: Initial reception and storage of
information, usually a syllable or word
(5) Concepts: Learning of simple object properties by
young children
.(6) Perceptual-motor skill: Learning to make continuous
tracking movements
(7) Problem solving: Discovering a principle which
achieves a stated goal.
Obviously, not many of these prototype learning situations, in and of
theMselves, sound much like "learning the facts of history from a textbook,"
or like "learning to demonstrate Coulomb's Law in the laboratory." Never-
theless, at some level of generality, these models all contribute to learn-
ing theory. For example, Short-term memory, the initial reception of
information, is an important part of every learning act. In considering
bow such principles apply to practical learning situations, it is mainly a
matter of deciding what can be assumed to122going on without a hitch, as
contrasted with what requires critical planning -.and arrangement.
The design of effective instruction, then, has these two areas of
knowledge to call upon.. Instruction needs to be arranged so that it will
bring about the kind of change in a student which is called learning, and
t'isrequires a Consideration of learning theory. In attempting to bring
Apout such a change,.the act of instruction is a matter of stimulating the
student in certain ways - - and here one has a choice of media to work with.
Putting ideas together from these two domains of knowledge can yield some
techniques and procedures of:instrUction which should make the process of
learning an optImallY effective one.
 Ig,219ns from L911,rialnaff2/92a
What specific suggestions about instruction can be derived from learn-
ing theory? As I have already noted, these are not self-evident. There
are many learning theo)?ies, and most of them are micro-theories, designed
to provide models of sonic relatively specific kinds of learning. Accord-
.ingly, a selection must be made among them, keeping in mind the purpose
of orientation toward the learning of young adults, or college students,
and with an awareness of the variety of media available.
I believe there are four different learning theorists who have pre-
sented ideas .of major importance to the design of instruction. These are
I

Neal, Miller, Skinner, Gagne, and Ausubel. I intend to describe these

ideas briefly here, before going on to elaborate on their implications.
It will be apparent that the suggestions of these theorists vary in their
: specificity, and I have ordered them along this dimension. Miller's ideas
are the most general, applying to a great variety of learning situations.
The specificity of suggestions increases progressively through the theory
Ausubel, who attempts to deal in a highly concentrated manner with the
problem of acquiring meaningful, organized knowledge.
N. E. Miller. Miller's views regarding the implications of learning
theory for, instruction are presented in'a volume of the Audio-Visual Com-
munication Review, entitled arlphic Communication and the Crisis in Educa-
tion (1957),. The four principles he describes are suggested by the words:
motivation, cue, response, reward. It is Miller's contention that an effec-
tive sequence of instruction, in any medium, must include provision for
these four condftions.
First, motivation: the student must want something. The motivational
effects of a "lesson" depend upon motivation which has already been learned.
To be most effective, the motivations aroused by instructional materials
must build upon those that are already in the life experiences of the stu-
dent. In other words, instructional materials cannot in themselves be
expected to generate brand-new sources bf motivation; but what they can and
should do is to capitalize upon, and add to, the kinds of motivations that
are already there. Various kinds of motivation may be called upon, includ-
.ng some presumably fundamental ones such as intellectual curiosity and the
desire to achieve. For students in college, learned motivations which form
t part of the individual's life goals, and which may exhibit themselves as
identification with admired people, as well as with choices leading to social
approval, are possibly of special importance.
 Second, there must be a cue: the student must notice something.
Mat....oials for instruction, whether verbal or pictorial, need to distinguish
the relevant, cues; Instructional materials are better to the extent that
they faciXJ:tate the discrimination of cues. Printed materials may do this
in a variety of, ways - - by varying type, by the use of color, but particu-
larly by means of their organization. Lecturers have a variety of ways of
distinguishing cues in auditory language - - by differences in loudness and
emphasis, and again by the organization of material. Pictorial presentations
obviously have used a variety of ingenious techniques of distinguishing cues
- by simplification, by the addition of pointers and markers, by the use
of color and contrast. The general point is that instruction will be enhanced
when the stimuli relevant to learning are readily discriminated by the stu-
A

dent.
Third, response: the student must do something. Many studies of learn-
ing have indicated the importance of student participation. Of course, the
doing may be a matter of internally conducted thinking or rehearsal. But
whatever form they may take, responses to instructional materials are an
4F ntial element in learning. Instructional effectiveness will be increased
the materials involve the student in doing something with
30) the extent that
his just-acquired knowledge.- transforMing it, applying it, using it.
Fourth, reward: the student must get something he wants. Various
techniques may be used to bring about satisfaction of this sort. Immediate
rewards are presumably more effective than delayed ones. Instruction needs
to reinforce the rewards learned in real life. For the student who is
motivated to solve problems and to achieve some learning goal, finding out
that.he has done well is an important reward. Instruction will be improved
in effectiveness to the extent that.some desired aim can be achieved, and
$

that knowledge of this achievement is given.

ObvioUsly, these four principles described by Miller are 'considered to
have highly general applicability to the design of instruction. They are
relevant to the learning of all kinds of students, and presumably to all
kinds of learning tasks. ,These principles may be put to work regardless
of whether one is considering the task of a first-grader in learning to
print letters, dr to the task of a graduate student in understanding a
solarly article. on Roman architecture.
--'14014111APIIIIPAIMINprimpupg

The generality of theso principles is also the key to their limitations

practical usefulness. To the skilled teacher or designer of instruction,
they seem obvious, and such a person would likely aver that he always uses
pjach prAnciples. Most instruction, in fact, could probably be shown to
incorporate these four principles in some degree. Even when one or another
is not strongly exhibited by instructional materials, it may be expected
that' an experienced learner (like a college student) will often arrange his
own learning conditions to include these principles. He comes with his own
motivation, he makes responses to what he reads, sees, or hears, and he
arranges his studying objectives so_ that some achievement will be noticeable.
Miller's principles are surely important to instruction, but it is question-
able whether they are often violated even in the most traditional instruction.
The views of Skinner on instruction are contained in a variety
Annars
of articles, particularly those on teaching machines (Skinner 1957, 1958,
1965).. Valuable analyses are also contained in books and articles by his
students (ilber1;,,1962; Green, 1962). At the most gene:cal level, it may
be' said that no great disagreement can be found with the principles of
ller. Skinner's analysis of instruction assumes that motivation must be
present, that the student must make a response, and that this response needs
to be rewarded, or "reinforced." The increased specificity of Skinner's
suggestions center a.. and the principle of stimulus control, or the ways in
which reinforcement may be used to establish both more precise and more
elaborate learnings by manipulation of the stimuli impinging on the learner.
In this sense, Skinner's views are most highly 'related to Miller's principle
about the importance of the cue in learning. It is possible to interpret
Skinnerian principles of instruction as a more extensive account of what
mupt be done to present cues (or stimuli) in such a way as to optimize learn-
ing.
Several relatively specific ways of controlling the learning process by
suitable sequencing of stimuli and reinforaxmipit: are suggested by Skinner's
theory. One is the principle of sh2.21. applicable to the learning of motor
acts_. As the individual practices a motor response of some sort, reinforce-
ment is given selectively so that the response which is originally only a
crude'copy of what is acceptable comes by a gradual process to be more and
Jre exact. Such a principle applies, for example, to learning to pronounce
.n unfamiliar language sound, such as the German umlauted u, or the French
uvular r. A second principle, somewhat similar, is that of successive
approximation of stimulus control, in which a response which is originally
 't

" Dmpted" comes to be given properly even when the prompt Alas been pro'gres-
,ively "faded." Initially, a student may need many contextual prompts, for
example, to remember what the Constitution says about the powers of the
President, but as he continues to practice recounting these powers, he can
do it without these extra cues. A third Skinnerian principle is chaininv,
which describes the conditions of reinforaement: by means of which a lengthy
procedure is learned. Essentially, the steps in the procedure, which might
be a computational procedure in mathematics, for example, are put together
in a step -by --step fashion, insuring that the final step is always connected
with the others which precede it (cf. Gilbert, 1962). ..

Thus it May be seen that the learning theory of, Skinner leads to some
1

relatively specific suggestions about the design of instruction. It gives

us practical procedures for shaping motor responses, for establishing dis-
criminations by successive approximations of stimuli, and for chaining
together the steps in complex procedures. For certain kinds of learning
tasks, these procedures are indeed specific and undoubtedly successful.
In my view, these principles are still only of general applicability
t the learning of certain other kinds of tasks, particularly concepts
and principles. For example, if one is concerned that a student acquire
an understanding of the principle of separation of powers as defined by
the Constitution, or an understanding of the principle of centrifugal force,
the notion of successive approximation provides only a very general prescrip-
tion fpr instruction. It says one must bring such behavior under finer_
stimulus control, but it does not specify how to do this. It does not say
how to select the stimuli which will accomplish this purpose. It seems to
me, therefore, that although some specificity about instruction in certain
tasks is definitely gained from Skinnerian theory, for certain others of
particular importance in college-level instruction, the suggestions remain
highly general.
gagr/e. The ideas of this theorist regarding the learning process are
contained in a book entitled The Conditions of Learning (1965), and their
applicability to instructional practice is discussed in a chapter of another
recent book, Indtruction: Some ContqlEporpry VitKpaints (Siegel, 1967). The
ggestionsto be derived from this view of learning are more specific for
instruction than are those previously described.
The first principle deserving emphasis is that of distinctive conditions
0

for different kinds of learning. Gagde distinguishes seven major kinds of

mental processing which are called learning, each of which has a different
 ,-9-

FiC of conditions for its optimal occurrence,. The seven kinds are called
signal learning (classical conditioning), S-R learning, motor and verbal
chain learning, multiple discrimination, concept learning, principle learn-
ing, and problem solving. He considers that the typical learning of young
adults, high-school and college students, may partake of any or all of these
types of learning, but that some are much more frequent than others in the
school environment. For example, certain motor and verbal chains may need
to be .learned in tackling a new foreign language, but these types of learn-
ing would probably never be encountered in courses in history, government,
or English composition, Most subjects in high school and collegd include
primarily the kinds of learning described as concept learning, principle
learning, and problem solving.
Although all types of learning may require certain general conditiOns
for their establishment, such as those of contiguM,, repetition, and rein-
forcement, emphasized by most learning theorists, the Lpecific conditions
for. establishment of concepts, principles, and rules are in addition to
.ese. Furthermore, they are distinguishable for each type: learning com-
plex principles through problem solving demands a different set of conditions
Alan does learning a new concept like "cell," "neuron," or "central nervous
system." The external conditions for each particular type of learning form
the basis for instruction. The internal conditions are retained capabili-
ties of the student which have been established by previous learning.
The second principle of importance for instruction may be called cumula-
tive le2.TALIK. This is the principle that the learning of any new capability
builds upon prior learning. According to this theory, there is a'specifiable
minimal prerequisite for each new learning task. Unless the learner can
recall this prerequisite capability (or some other which can serve the same
purpose), he cannot learn the new task. As a very simple example, unless a
learner can recall how to factor numbers, how to divide, and how to multiply,
he cannot learn to find a lowest common denominator, and nus to add fractions
This principle has a deceptive simplicity about it, and may readily be dis-
missed as either obvious or trivial. In actuality, it is neither. It does
not say, before the learner undertakes to 'learn how to add fractions, he must
nave "had" or been through" the factoring of numbers. Instead, it says he
must have mastered and must be able to recall the factoring of numbers in
rder for the desired leas wing to take place at all. This principle is con-
sidered to have broad applicability to the learning of principles; whether'
they bd thd origin of the American Revolution, the generation of induced
 -10-

es" ctric current, or the constancy of perceived size. In all of these"'

Instances, there are specific minimal prerequisite learnings, before the
new learning task is undertaken.
Ausubel.The views of this learning investigator may be sampled in
The Psynclogy of Meaningful Verbal Learning (1963), and also in an informa-
tive chapter in the book Instruction: Some Contemporary Viewpoints (Siegel,
1967).
Ausubel insists, first of all, that school learning is me?ningfalearn-
ing and that this process is distinctly different from what is usually called
rote learning. Thus he comes to grips directly and specifically with the
learning of facts and principles, and is not particularly concerned with
other forms of learning such as motor and verbal chains. In this theory,
the most important principle is called subsumption. Meaningful learning
takes place, according to this theory, when a new idea is subsumed into a
related structure of already existing knowledge. The result of this process
is the acquisition of a set of new meanings.
There are a number of implications of this view for instructional prac-
.t ,e. For example, one is the importance of providing the learner with a
,meaningful structure, before he attempts to learn a new principle - - an
organizer, which bears a logically superordinate relation to what will be
ldarned. Putting this in a somewhat oversimpl3fir3d form, it means that if
the learner is expected to learn about coal and oil and gas, one must tell
him ahead of time that he is going to learn about "the different forms of
fuel." A second principle is that any subject should,be presented by
progress five. difierentiatjon of content, the most general and inclusive ideas
first, and then the more detailed and specific ones. Ausubel states that
although this seems a self-evident principle, it is rarely followed in
actual teaching procedures or in textbooks.
Still a third principle of importance is called consolidation. This
means the insistence on mastery of ongoing lessons before new material is
introduced. This proposition is at leadt highly similar, if not the same,
as Gagne's principle of cumulative learning. Another Ausubel principle of
great importance would, seem to be integrative reconciliation. By this he
mans that new ideas, once introduced, need to be deliberately related to
0.Ld ideas, s.ignificanu similarities and differences pointed out, real or
.pparent inconsistencies reconciled. Again; Ausubel finds this a practice
followed scarcely at all by textbook writers.
 fi

These principles add up to a pretty strong specification of how

Instructional materials should be organized and presented for most effec-
tive learning. While one finds only very general guidance for the con-
struction of prpgrams of instruction, texts, or educational films by follow-
ing such principles as Miller's, it is evident that Ausubel's principles are
pretty specific. They tell an instructional designer what to do first, what
sequence of ideas to follow, what to do to insure remembering, and what kind
of outcome to expect. Note that I do not maintain that Ausubel's theory is
entirely correct - only a good deal more experimentation will determine
that. But his ideas lead to very concrete suggestions about how to, conduct
instruction.
Here then we have four theories of learning, each of which has some-
thing to say about how to design instruction. Virtually no instructional
materials, texts, or films in existence today have deliberately been pre-
pared on the basis of these principles. Today's instruction simply does
not reflect these principles, but appears instead to be based upon an older
-et of principles derived from quite different considerations. Could in-
structional materials be designed to take these principles into account?
I see no reason why this could not be done. It would be an expensive under-
taking, even to design a single course this way. I am unable to estimate
cost effectiveness -- indeed this may not even be possible until someone
has tried -Co do it once.
Otherwise, the kinds of principles I have been talking about can even
now be put into effect in at least a partial fashion by, first, the instruc-
tor, and second, the student. For example, the instructor can use the prin-
ciple of organizers, and the principle of integrative reconciliation,'even
though he may not be able in any immediate sense to rewrite the textbook or
redesign the television lesson. The student is also able to put many of
these principles into effect himself. In fact, it seems probable that what
is meant by a sophisticated learner, as opposed to a novice, is one who
imposes, his own organizations on presentations of material, arranges his
own distinctive conditions for learning different kinds of tasks, carries
out his own integrative reconciliat4on of new and old ideas. Learning to
o these kinds of intellectual activities, to carry out these kinds of
strategie3,'may represent an educational goal of more fundamental importance
han the learning of any particular set of facts, rules, or principles. 0
 -12-

L.erning and the Individual

This possibility of the learner's contribution to his own learning sug-
gests an even broader theme than any which has been specifically defined by
learning theories. Perhaps it may become the most general principle of all.
It may be said, surely, that the great majority of modern studies of learn-
.ing, of a variety of types, provide an accumulating body of evidence for
this principle: Learning and remembering require the imposition of an
active intellectual process by the learn'er on the material presented to his
senses. One simply cannot account for learning by specifying only what is
presented and the level of "intelligence" of the leanner. -Apparently, some
specific sort of processing is all'ays contributed by the learner himself.
This kind of processing is given various names, in various experimental set-
tings. For example:
1. In studies of rote verbal learning, it is typically called mediation
(cf. Jenkins, 1963). Learning to associate a nonsense syllable like DEP with
a nonsense syllable like RIV has been shown typically to involve the contribu-
tion of a linking mediator by the learner. (In this particular instance, it
it be a word like "deprive," or two words like."deep river").
2. In still other studies of memory, the process may be referred to
as coding (cf. Melton, 1963). Investigations of short-term retention of
small verbal units are generally considered to reveal important facts about
the "intake". portion of the learning process. Here, it is found that a
-single syllable like XQR is not retained as well after one A?esentation as
a syllable like NER, nor as well as a word like TOP. In fact, the single
syllable XQR is retained no better than three short words. The suggestion
is that something is done to these units before they are stored. They are
first coded.
3. Investigations of the learning of concepts by children (cf. Kendler,
1964) provide other sources of mediational processes. It is found that chil-
dren of four years of age cannot shift readily from one concept (like "black")
to its opposite, whereas seven-year olds can. By inference, this is because
the older children have a greater store of "mediators" to apply to this
reversal situation.
4. Many studies of problem solving in young adults (cf. Gagnb, 1964),
a variety of tasks, have emphasized the impor'ance of prior knowledge,
an organization based upon prior knowledge (cf. Katona, 1940), to the
successful solution of problems. Problems are solved when the learner is
able to bring such an organization, which is already available to him, to
bear upon the task at hand.
 :(
-13-

5. Studies by Rothkopf and his associates (cf. Rothkopf, 1966), -have'

,own the impOrtant effects on learning and retention resulting from the
introduction of questions into textual passages, even when the questions
ere irrelevant to what is ultimately tested as having been learned. In
other words, the inference is that the learner applies to the learning
task a complex set of behaviors which may be approximately summarized as
a "set to remember."
'These are only a few of the many lines of evidence showing that new
learning cannot be adequately accounted for in terMs of what is presented
to the learner. In addition to these external stimuli, a very important
part of the process is contributed by the learner himself, or more specifi-
cally by what is stored in his central' nervous systeffi, Furthermore, it is
fairly clear that the coding or mediation done by the learner depends upon
his particular store of past experience. The particular way the learner
codes a presentation is peculiar to him, and not shared by other individuals.
.The mediation of leamInz_is
The implication of these findings is quite clear. So far as theories
c the learning .process are concerned, the learning of any set of materials
lepends importantly upon individual contributions from the learner himself.
Learning is an individual matter. In a fundamental sense, it is determined
by what the learner does, and not by what the material does or what the
teacher does. One can even go a step farther, in drawing implications for
education. If one is concerned about hOw to make learning efficient, the
focus of emphasis must be the student. The design ofefficient conditions
,for learning demands that learning be conceived as an individual matter.
Novi, there are conflicting views on this question. Some psychologists,
looking at,the educative process as typically involving a teacher and a
'class, have emphasized the teacher-student interaction, or what is sometimes
called the teaching-learning process. Jackson (1966), for example, distin-
guishes between teacher-student dialogues which are private (as in a tutor-
ing situation), those which are public (as in a classroom), and those which
might be called semi-private, in whiCh the teacher works with a single stu-
dent while others engage in some other activity. He correctly notes that
learning theorists have seldom if ever contributed to an understanding of
t public teaching situation. Another theorist about teaching is Thelen
1967), who has carried out a series of most interesting studies of teacher-
/
student'interactions, seeking ways of find a "fit" between teachers and groups
of students. The absence of change in the school's output resulting from
changes in administrative procedures such as class size, team teaching,
ai .4.ity grouping, as well as instructional procedures like discussions vs.
'.ectures, leads Stephens (1967), to the conclusion that as long as a teacher
has a strong interest in his subject, what he does is relatively unimportant.
For many legitimate purposes, there is surely much to be gained by study-
ing the activities of the teacher, and theorizing about how he interacts or
should interact with, a learner. But such studies can tell us little about
how learning occurs, or how to make it efficient., 1f wewish to find out
about learning, we must begin and end with the human individual who is the
learner. We must, in other Words, find out what the learner is like, what
he needs to know to begin. the learning process, and what he needs to do to
carry it out. The ,site of learning is not in a group, nor is it in a. rela-
tionship between instructor and student. The site learning is the.indivi-
dualls central nervous system. For this fundathental and unarguable reason,
learning. is individual. Efficient instruction, Is designed for the individual
learner.
.The recognition of the individual character of learning need not blind
u' to some of the necessities of public communication, with both teachers
.and other students. Schools and colleges are concerned with the transmis-
Ion of public knowledge. There is, of course, such a thing as strictly
private knowledge, as for example that exhibited in artistic accomplishment.
But the schools cannot transmit this private experience, by definition. The.
communications of knowledge become refined,' sharpened, and clarified by
public discussion. In schools, therefore, public discussion serves the same
highly essential purpose as it serves in other settings in the larger commun-
ity. In a university setting, there is a great deal of public discussion,
and it is highly important for the clarification and refinement of the "mes.r
.sages" that are to be transmitted. Often, .discussion takes place in a class-
-room, among students, and between students and teachers. Much discussion
takes place among faculty members., And obviously, a great deal takes place
outside of class among students. I believe discussion is a highly impomant
Tart of school learning. Unfortunately, it must be said that we have no .

theory as yet of the role of discussion in learning. Such a theory, it may

be expected, will not be opposed to a theory of individual learning, but will
E plement it.
Instruction and the Individual Learner
It is possible, then, to bring to bear upon the design of instruction
some principles of learning theory. These principles range from those which
are quite generally applicable to all forms of learning to those which apply
 sylifically to the learning of concepts and principles of the sort Which
iharacterize the bulk of knowledge taught in the schools. In addition,
modern studies of learning suggest the clear implication that some idiosyn-
cratic processing of information is done by the learner. This provides a
fundamental reason for viewing aearning as an individual process, and strongly
suggests that individualized instruction represents the route of efficient
learning. If arrangements for individual learning are not deliberately made
by the system, they presumably will be made by the learner hithself. In doing
this, he will presumably use whatever media are available, although some may
be better adapted for some purposes than are others.
The "arrangements" of the external environment for purposes of efficient
learning are what constitute the events of instruction. One should not lose
sight of the fact, though, that learning in the sense used ,here also includes
remembering and transfer of learning, since it is these less immediate out-
comes that are the true concerns of an educational system. Assuming that
these are included, what ara the events of instruction that must take place
if, order for learning to occur?
In framing an answer to this question, I should first point out that
.ccording to GagAels (1967) conception, the conditions of instruction differ
with the type of learning being undertaken. Thus one does not design Instruc-
tion on using a key-puneti machine to be the same In its formal characteristics
as instruction on how the mechanism of akey-punch machine operates. Or, in
learning a foreign language, one Jees not design instruction on pronouncing
words to be the same as instruction on understanding spoken sentences. There
are some important distinctions here which should not be overlooked. Hewever,
for purposes of the present paper, I shala not elaborate them further. In-
stead, I shall speak only about the events of instruction applicable to the
learning of principles, including facts, generalizations, and rules.
What appear to be the most important events of instruction are the
following:
1. Gaining and maintaining attention. Obviously, in order for learntng
to occur, attention must be attracted in the first place, and then maintaiyled
Many of the stimulation, conditions that'attract attention have been known
fo?, a long time, including such things as change, novelty, app, al to dominant
inuerests. Concerning the maintenance of attention, we know somewhat less.
I. jothe clarification has surely been gained by Travers' (1964) demonstraticn
that we only attend to one thing at a time, regardless of how many media
channels may be bombarding us. Presumably, maintaining attention is a matter
of achieving a set, related to one or more individual voals, which makes the
1 rner return again and again to the task at hand. Manipulating external
Stimuli is probably ineffective over the long pull, and one must instead
seek ways of reinforcing, the motivational state of .the learner.
2. Insuring recall of previously acquired knowledge is another import-
ant function of instruction. We have seen that recall of prior knowledge is
considered an essential condition of learning by both Gagde and Ausubel.
When the learner, undertakes to learn something new, he must first be reminded
of what he already knows which is relevant to that learning.
3. Guiding the learning is done in instruction by verbal or pictorial
material that provides "cues" or "hints" to new principles, usually without
stating them fully in verbal form. In part, the "organizers" mentioned by
Ausubel perform this instructional function. In part, it is done by ques-
tions, as Rothkopf's work illustrates. The skilled self-learner, of course,
rovides his own questions.
4. Providing feedback to the learner on his accomplishments is another
function of instruction. One of the surest ways, it seems to me, is by
0-fining the objectives of instruction clearly, to the learner, so that he
will become aware immediately when he has attained each specific goal. Again
bhe skilled learner may usually do this himself. Textbooks and other media
often seem to neglect badly this essential instructional function.
5. Establishing conditions for remembering and transfer of learning
would surely be counted as one of the essential functions of instruction.
For pUrposes of transfer, there needs to be a carefully designed series of
problems to which application of the newly learned principle is made. Prob-
ably also having this function is the process Ausubel calls "integrative
reconciliation," in which new ideas are compared and contrasted to related
ones previously learned. For remembering, there needs to be provision for
spaced-review, which has often been shown to be an effective technique (cf.
Davis, 1966, pp. 55-71).
Finally, they should be mentioned still another instructional
6.
function, often neglected. This is the assessment of outcomes. The out-
comes of learning and remembering need to be assessed frequently. The admin-
istration of a final test or examination for purposes of determining a grade
!ems often to be a way of consolidating an onerous task which because of its
unmanageable scope ends up avoiding the very assessment that should be done.
Learning of the specifics needs to be assessed, perhaps more so than learn-
ing of the generalities. The five-minute daily or weekly quiz has much to
recommend it. For the skilled learner, this function can often be performed
To-i 4-1-1 QiinnPqg her hirnqPi'r. T1OL to test onzoolf is indeed a highly
 ..17-

s( dsticated thing to do, and instructional materials should provide

such. help as passible in thli function.
instruc-
There are, then, these six major functions that take place in
It may be noted that learning theoz;y does not in and of itself,
tion.
variety of speci-
say exactly how these are to be put together in the great
tells us,
fic instances to' which they are applicable. What learning theory
will occur,
is that when certain of these conditions are present, learning
Beyond such
and when certain ones are not present, learning is improbable.
this
theory there must of course be both technology and artistry, whether.
teacher..
be exhibited by the textbook writer, the film-maker, or the master
effective techniques
And to a considerable extent, at least, we should expect
of self-instruction to be present in the young adult.

What Can Media Accomalish?

readily
It can readily be seen that most media of communication can
be performed by
perform most of these instructional functions. They
pictures, by printed language, by auditory language, or by a combination
is not the message.
( media. So far as learning is concerned, the medium
uniquely adapted to_ per-
No single medium possesses properties which are
Instead; they all
foFm one or a combination of 'instructional functions.
The arrange-
perform some of.these.functions well, and.some not so well.
instruction,
ment of instructional conditions is still the key to effective
regardless of the medium or media employed.
considering
One key to the question of which media is to be found by
A properly
the learning task, that is, the objectives of the learning.
nature of stimuli to
defined set of objectives provideS information on the
Consider
which the learner is expected to respondafter he has learned.
, ;

4
a .few examples: .

1. An objective in a course In physics might be, "demonstrating Ohm's

reeStance in a electric cir-
.Law." If one expects the student to show how
be consider.-
cuit varies with the current and voltage, there would seem to
for
ablejastification for using actual objects and events as the medium
,

in a laboratory.
instruction. In other Words, one'might sot up instruction
such actual objects
the student has sufficient prior acquaintance with
and events, a pictorial presentation may perform the same functions.
2. An objective in a course in English might be, "editing composed
clarity of
written paragraphs for correctness of structure and optimal
initially are in-
expression." Obviously, what has to be presented here
to be the medium.
correct and non optimal paragraphs. Printed language has
 1

to

-18-

He ver, it may be of considerable importance in such an instance to arange

'or frequent and prompt feedback to the learner as he makes his corrections.
Thus one might choose to have a teacher convey this feedback in the presence
of printed language given in a text or, projected on a screen.
3. In a foreign language course, an objective might be, "making appro-
..

priate responses containing personal biographical information to questions

asked by a speaker in the foreign language." Here again, the medium required
is quite evident - - it is auditory language. The learner, must be presented
with these questions in an auditory form, and the printed form will not be
an adequate substitute.
Consideration of these examples, and others like them, leads to the
following generalizations concerning the use of media for instruction. These
seem to me to be more or less self-evident principles with which one must
begin to think about media. They are not "the answers," but merely the
basis for further investigation of the uses of media.
1. First, no single medium is likely to have properties that malce_it
best for all purposes. There.is, so far as we know, no special magic in
and particular medium.
2,Second, the most important single criterion for a choice of medium
is often the nature of the learning task itself 4' - that is, the objective
of the instruction. If the learner is going to respond to real objects,
these need to be used at some point in instruction. If he is going to re-
spond to auditory'language, then this form of communication needs to be
used at some point in, his instruction. However, it should be noted that
this criterion doesn't solve the whole problem, by any means. The reason
is that for many objectives, one medium is as appropriate to the task as
another'. For example, the principle relating the sides and hypotenuse of
a right triangle can be presented in printed words, in mathematical symbols,
or in diagrammatic pictures. Or, the events'leading up to the Boston Mass-
acre can be described ina printed text or shown in dramatized pictorial
form. In these instances, nothing in the instructional objective itself
provides a clue as to which medium will be best.
3.
Third, when one considers the six functions of instruction (control-
) g'attention, stimulating recall, etc.) previously mentioned, it is evident
that any given medium may perform one of these functions best at a given time
during a period of instrucing, while another medium may perform an instruc-
tional function best at-another time. That is to say, the precise answer to
the question of "which medium" is not to be found by matching courses with
pPr1iA. nr Pvtz.n tnninR with media. but rather in matching specific irls;ructiona_
 -19-
/
functions with media. Within a given topic, for example, attention might
)est be maintained by the introduction of pictures, whereas guiding learn-
ing might best be accomplished by printed verbal instructions, and feed-
back might be best: performed by auditory language. This line of reasoning
: Is develope.d more fully by Briggs et al (1967), in a monograph on Instruc-
tional Media. When one chooses a particular medium for a.whole course, or
even for the, development of an entire topic, one is usually making a judg-
ment that such a medium will be best suited "on the average" for the various
instructional functions it must perform.
4. Finally, there is another suggestion to be derived from thesecon-
siderations about the instructional functions of media. It may be that the
most striking effects of instructional planning are to be sought in.various
. combinations of mediaj where each may perform a particular function best.
This does not mean reverting to the idea which Travers' (196) work calls
into question, that simultaneous auditory and visual presentations are
superior to either alone. What it means instead is that any given medium
1.....ght be used alternately with others over relatively short periods of
instructional time.
tk
Consider, for example, an instructional situation in which the student
reads from a printed text and responds to it by writing problem answers.
When the occasion demands, pictures or diagrams are presented to perform
the functions of stimulating recall and guiding the learning. Now, as the
student works along in this fashion, every so often, when a new subtopic
is to be introduced, or special emphasis isto be given, a taped auditory
message is introduced, having the primary purpose of con rolling attention.
Frequent questions are included in the printed text for self-assessment,
and feedback is also provided in an auditory form. What would be the effec-
tiveness of this kind of combination of media?
I do not know the answer to this question, and there is no research to
provide it. Yet this kind of instructional Arrangement, only roughly des-
cribed in this example, may hold the key to effective instruction, particu-
larly the sort of instruction which depends' upon the individual to do a
1.rge part of the establishment of learning conditions for himself. Obvi-
ously, a good deal of testing of practical alternatives is needed before
we can feel confident about the outcome of such plans for instruction.
I have been led in this paper to consider first how learning theory °

relateS to the practical events of instruction. There is little doubt that

this relationship can be demonstrated. Depending; upon which learning theory
1

-20-

chooses, the suggestions for practical application to instruction are

lone or less specific. And running through all theories is the theme that
'learning is, after all, an individual matter, in which essential idiosyn-
cratic elements must be supplied by the learner himself.
As a practical matter, the events of instruction encompass more pro-
cpsses than are included in learning theories themselves. Instruction
involves gaining and controlling attention, stimulating recall, guiding
the learning, providing feedback, arrc4liging for remembering, and assessing
outcomes. It is these functions that are performed by various media of
instruction, and to a considerable degree by the learner himself. One
should not expect, I think, to find that a single medium is best fitted to
do all of these things. Instead, it seems likely that carefully designed
combinations of media may be required to achieve the kind of instruction
that is most effective, and which at the same time exploits the properties
of media td fullest'advantage.
 -21-

References

Ausubel, D. P. The PszahologLof Meaningful Verbal Learning. New York:

Grune and Stratton, 1963.

.Ausubel, D. P. A cognitive-structure theory of school learning. In L.

Siegel (Ed.), Instrucfton: Some Contemporary Viewpoints. San Fran-
4.
. cisco: Chanler, 19-67, pp. 207-257.

'Briggs, L. J;, Campeau, P. L., Gagne, R. M., and May, M. A., Instructional
Media. Pittsburgh: American Institutes for Research, 0677-0137Eph
No. 2T.
Davis, R. A. Learning in the Schools. Belmont: Wadsworth, 1966.

Gagde, R. 'M. The Conditions of LearnLag.. New York:' Holt, Rinehart, and
Winston, 196.
Gagde, R. M. Problem solving. In A. W. Melton (Ed.), Clate,:gorig,.1(2fHuman
LearnIng.. New York: academic Press, 1964.

Gavle, R. M. Instruction and the conditions of learning. In L. Siegel (Ed.

Instruction: Some ContempoLary_yiewpolnIs. San Francisco: Chandler,
pp. 291-313.

Gilbert, T. F.. Mathetics: the technology of education. Journal of Mathetics

1962, 1, 7-73..

Green, E. J. The Learning Process and Programmed Instruction. New York:

Holt,.Rinehart and Winston, 1962.

Jackson, P. W. In The_WayLTeaching
The way teaching is. Report of the :

Seminar on Teach. Washington, D. C.: National Education Association,

1966, pp. 7-27.

Jenkins, J. J. Mediated association:. paradigms and situations. In C. N.

Cofer and B. S. Musgrave (Eds.), Verbal Behavior and Learning. New
York: McGraw-Hill, 1963.

Katona, G. Organizing and Memorizing.. New York: Columbia University Press,

1940.

.Kendler, H. H. The concept of the concept. In A. W. Melton (Ed.), pateRpries

.of HuMan Learninp. New York: Academic Press, 1964.

.Melton, A. W. Implications of short-term memory for a general theory of

memory. Journal of Verbal hearnincr and Verbal Behavior, 1963, 2, 1-21.

Melton, A. W. patepories of Human Learning. New York: Academic Press, 1964.

_Miller, N. E., et. al. Graphic Communication and the Crisis in Education.
Washington, D. C.: Department of Audio-Visual Instruction, National
Education Association, 1957. (Audio-Visual Communication Review, Vol.
5, No. 3).
A

-22-

Rothkopf, E. Z. Some theoretical and experimental approaches to problems in

written instruction. In J. D. Krumboltz (Ed.), Learning and the Educa-
tional Process. Chicago: Rand-McNally, 1965, pp. 193-221.

Siegel, L. (Ed.), Instruction: Some Contemporary Viewpoints. San Francisco:

Chandler, 1967.

Skinner, B. F. Verbal Behavior. New York: Appleton-Century-Crofts, 1957.

Skinner, B. F. Teaching machines. Science, 1958, 121, 969-977.
Skinner, B. F. Reflections on a decade of teaching machines. In R. Glaser
(Ed.), TeachinE Machines and Programmed Learning: II. Data and
Directions. Washington, D. C.: National Education Association, 1965,
pp. 5-20.

Stephens, J. M. The Process of Schoolinrf. New York: Holt, Rinehart, and

Winston, 1967.

Thelen, Classroom Grouping 1 Teachability. New York: Wiley, .1967.

Travers, R. M. W., et al. Research and_The= Related to Audiovisual

Information Transmission. Salt Lake City: Bureau of Educational
Research, University of' Utah, 1964.