Achievement Motivation: history
• Murray’s Explorations in Personality
Approach Motivation
The theory of Achievement Motivation
Murray’s Explorations in Personality
• Intense study of small set of subjects from
many different perspectives
• Conceptual identification of needs
• Development of Thematic Apperception
Test
– Needs driving perception and production
McClelland and Need for
Achievement
• N-ach and the achievement of nations
• Cultures with a high need for achievement
(rather than some other need) will strive to
overcome obstacles (other nations?)
– Greek civilization and Greek literature
– N-ach in children’s primers and later economic
growth
– Teaching n-ach as a means for development
• McClelland and the Need for Achievement
• Atkinson and theory of risk preference
– Static
– Dynamic
• Weiner and attribution theory
• Reinvigoration: Elliot and Thrash
Need for Achievement
• Desire to approach problems involving
challenge and effort
• Joy in success when over coming obstacles
• Analogous to a hunger
• “The little engine that could”
– “I think I can, I think I can, I think I can”
Issues in measurement
• Projective measurement
– Can’t trust self reports of motivations
– Ambiguous stimuli will lead to interpretations
in terms of motives
• Hunger and interpretation of ambiguous slides
• Achievement and stories
– “grubby graduate student” versus “professor”
1

Issues in measurement: II
• Weiner’s 3 points:
– TAT is the best way to measure motivation
– TAT is the worst way to measure motivation
– People who use TAT believe 1, people who do
not believe 2
Static theory of risk preference and
achievement motivation
• Achievement motivation: the joy of success
• Approach motivation
• Atkinson’s theory of risk preference (1957,
1964)
– An expectancy value theory of motivation
– Contrasted to drive models of Hull, Spence
• Tendency to approach = Value * Expectancy
Value = Motive * Incentive

Specific model for achievement Fear of Failure: the pain of failure

• Expectancy = subjective probability of success
• Motive = Individual’s need for achievement
• Incentive = difficulty = 1- probability of
success
• Conclusion for achievement motivation
– Ts = Ms * Ps * (1-Ps)
– Implies that motivational strength is quadratic
function of probability of success
• Fear of failure -- test anxiety?
• Fear of failure and general avoidance motivation
• Specific assumptions for fear of failure
– Expectancy of Failure = Pf = 1-Ps
– Motive to avoid Failure = fear of failure = Maf
– Incentive to avoid failure = - easiness = - Ps
– Taf = Maf * (Pf) *(-Ps) = Maf * (1-Ps) * (-Ps)

Resultant Achievement Motivation Tendency by Ps by Ms and Maf

0.3

0.25

• Resultant tendency = tendency to engage in

0.2

a task for success + tendency to avoid

0.15

failing (negative) + extrinsic tendencies

Resultant tendency

0.1

• Tr = Ts + Taf + Text 0.05

• Tr = Ms * Ps * (1-Ps) + Maf *(1-Ps) * (-Ps) 0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

• Tr = (Ms-Maf) * (1-Ps)*(Ps) -0.05

-0.1

-0.15
Probability of success

Approach Avoidance Resultant

2
 Motivation, risk preference and
Tests of original theory
persistence under failure
• Motivation and risk prefence: the ring toss Easy (p = .7) Hard (p = .05)
– Hamilton
– Heckhausen
High Nach 6/8 2/9
– Although inverted U, did not peak at .5
difficulty (Low Maf)

Low Nach 3/9 6/8

(High Maf)

Feather, 1964

Tendency by Ps by Ms and Maf:

Revisions to Atkinson Theory one trial 0.3

0.25
• Raynor and the concept of future orientation
– Life is not a ring toss - tasks are contingent 0.2

– Probability of success at event i = ∏pi = p 1*p 2 …pn 0.15

Resultant tendency

– Consider a freshman starting psychology with p = .9 0.1

• 110 201 205 215 301 398 grad MA PhD

job tenure full 0.05

• .9 .81 .73 .66 .59 .53 .48 .43 .39

.35 .31 .27 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

– Tendency to engage in a task = sum of tendencies for tasks -0.05

contingent upon that task

Trn = ∑(Ms-Maf ) * Psic * (1-P sic) + Text -0.1

-0.15
Probability of success

Approach Avoidance Resultant

Contingent Paths: Total Tendency

Contingent Paths: Preference as a
for 3 trial path
function of probability 3 trials
0.3
0.8

0.25

0.6

0.2

0.15 0.4
total action tendency
action tendency

0.1

0.2
0.05

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
-0.05

-0.1 -0.2

-0.15
Probability of success per trial
-0.4
Approach Avoidance Resultant probability of success

Approach Avoidance Resultant

3
 Contingent paths: Evidence for
Implications of contingent paths
Raynor’s hypothesis
Study1 Motive to Low High
• High achievers should set distant goals
achieve
– Low achievers should set immediate goals
Importance to future
• Preferences for task difficulty should vary
High (major) 2.9 3.4 as a function of number of outcomes
Low (distro) 3.0 2.6 contingent upon particular task outcome

Study 2 High 3.0 3.5

Low 3.4 3.4

Further explorations: curvilinear Class Performance and Test Scores:

models A simple model
• Does task performance vary as a curvilinear • Assume variation in ability 1-5
function of task difficulty • Assume motivation in class varies 1-4
• Assume motivation in test situation = resting
• Is it overachievement or under (class) + 1
performance? • Assume efficiency varies as inverted U of
motivation (max at 3)
• Assume test performance=ability*efficiency
• Assume cumulative performance
=ability*efficiency* time spent

Test and Class Performance

Class and Test Performance Motivation in Efficiency Performance
Ability Class Test in class on test Time Spent On test in class2
1 1 2 1 2 1 2 1
2 1 2 1 2 1 4 2
Test vs. class performance
3 1 2 1 2 1 6 3
50 4 1 2 1 2 1 8 4
45
5 1 2 1 2 1 10 5
1 2 3 2 3 2 3 4
40
2 2 3 2 3 2 6 8
35 3 2 3 2 3 2 9 12
4 2 3 2 3 2 12 16
30
Class Performance

5 2 3 2 3 2 15 20
25 1 3 4 3 2 3 2 9
20
2 3 4 3 2 3 4 18
3 3 4 3 2 3 6 27
15
4 3 4 3 2 3 8 36
10 5 3 4 3 2 3 10 45
1 4 5 2 1 4 1 8
5
2 4 5 2 1 4 2 16
0 3 4 5 2 1 4 3 24
0 2 4 6 8 10 12 14 16
Test Performance
4 4 5 2 1 4 4 32
5 4 5 2 1 4 5 40

4
 Dynamic theory of achievement Trial to trial carryover effects
• Recognition of inertial properties of • Weiner and Schneider carryover and
motivation interpretation of success and failure
– Motives persist until satisfied – Success and failure on verbal learning tasks
– Lewin and the “Herr Ober effect” – Anxiety inhibits performance on hard tasks
– Zeigarnik and the motive for completion – Anxiety facilitates performance on easy task
• Completed tasks – T res = Tapp -T avoid
• Uncompleted tasks

Weiner and Schneider, 1971

Drive Theory Predictions
Drive vs. Cognitive Theory
• Prior work using Drive Theory had suggested that
high anxiety interferes with difficult but facilitates sEr = sHr *(D+K) Strong Habit
easy tasks.
– (Very well established result with >25 replications)
– Based upon Drive theory interpretation that Anxiety sEr
increases drive and that the Evoked response is a
function of Drive X Habit Weak Habit
– Assume that Easy => Correct Response is dominant,
Hard, => incorrect Response is dominant
– Typically use serial anticipation

Low Anxiety
Drive -> High Anxiety

Weiner and Schneider, 1971 Weiner and Schneider, 1971

Trials to Criterion by Feedback
• Task: Learn 13 CVC trigrams 40

Easy List: high between item differentiation

35
e.g. PAK, BIM, MOT Difficult

Difficult list: low between item differentiation 30

e.g. HOV, VOV, RIV, MIV 25

Trials to Criterion

Lists presented as serial anticipation (implicit feedback?) 20

Easy
Subjects were high and low resultant Achievement 15
Motivation (Nach - Naf)
10
Feedback - list is (easy/hard) you are doing better/worse
than others 5

0
Success Failure

High hard Low Hard High Easy Low Easy

5

Locke and Goal Setting
• Thorough review of goal setting effects:
– The harder the goal, the higher the output
– Hard tasks lead to more effort than easy tasks
• This is inconsistent with Achievement
motivation theory that effort is greatest for
moderately difficult tasks
Revelle and Michaels: steps
towards dynamics
• How to reconcile the simple try harder the
harder the problem (goal setting, see Locke)
model with Atkinson model
• Hard tasks take longer to complete and if
there is carryover from trial to trial, then
motivation should accumulate

Expected Effort as a function of

Steps towards dynamics
trial and probability of success
0.45

• Effort on trial 1: Ms-Maf*(Ps)*(1-Ps) 0.4

2nd trial
• Effort on Trial 2 is a function of outcome of 0.35

trial 1: 0.3

0.25

– If success on trial 1, then effort T2 = T1

Effort

0.2
1st trial
– If failure on trial 2, then motivation from trial 1 0.15

carries over to trial 2: Effort T2 = T1 + carryover 0.1

– Assume perfect carryover T2 = T1*p + 2T1*(1-p) 0.05

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Probality of Success

1 trial 2 trials

Steps towards dynamics Carryover (3 trials)

• Effort on trial 1: Ms-Maf*(Ps)*(1-Ps) Trial 1 T1=p*(1-p)
• Effort on Trial 3 is a function of outcome of
trial 2: outcome p(success)=p P(failure)=(1-p)
– If success on trial 2, then effort T3 = T1
Trial 2 T1 2* T1
– If failure on trial 2, then motivation from trial 2
carries over to trial 3: Effort T3 = T3 + carryover outcome p(s)=p2 f=p*(1-p) S=(1-p)*p F=
– Assume perfect carryover (1-p) 2
Trial 3 T1 2* T1 T1 3* T1

6
 Perfect carryover 1-3 trials What if there is less than perfect
carry over from trial to trial?
0.5

0.45
• Motivation carries over from trial to trial,
0.4 but some effort is expended so there is not
0.35
perfect carryover.
0.3

• Consider 90, 80 and 70% carryover

Effort

0.25

0.2

0.15

0.1

0.05

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Probability of success

Series1 Series2 Series3

Effort and consummation

repeated trials Dynamics of Action: Approach
Atkinson and Birch, 1970
0.8

0.7

0.6
• Action Tendencies as latent needs
Expected effort - repeated trials

0.5
• Instigating forces -- situational stimulation
and individual sensitivities
0.4

• Consummatory forces -- need satisfaction

0.3

0.2
• Change in action tendencies = f(instigating
forces - consummatory forces)
0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
probability of success

no carryover 70% 80% 90%

Action tendencies over time

Dynamics of Action
F=1 or 2, c = .1 or .2
Atkinson and Birch, 1970 25

• Action Tendencies increase as a function of

20

instigating forces, decrease as a function of

action. 15
Action Tendency

– dT = F (if not ongoing)

– dT = F - cT (if ongoing) 10

– Stable state occurs when dT = 0 <=> T=F/c

• Actions with greatest action tendency will occur 5

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Time

F=1 c = .1 F= 2 c = .1 F=2 c =.1 F=2 c =.2

7
 Incompatible actions over time
A dynamic dinner party
35
16
Lagged consummation
14
30

12

25

10

20
hunger

Series1
8
Series2

15

10
4

5
2

0
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
time

Incompatible Action tendencies

25
Ongoing decays Incompatible actions over time,
the problem of “chatter”
25

20

20

15

15
Action tendecies

Series1
Series2
Task 1
Task 2
10
10

5
5

0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
Time

Avoidance and Inhibitory Inhibition and resultant action

Motivation -- Negaction tendencies
• Negaction tendencies inhibit behavior • Resultant action tendency = T -N
• Inhibitory forces increase negaction • Resultant action tendency will grow if not
• Resistance forces decrease negaction ongoing
• Dn=I-rN <=> N -> I/r at limit • Example of bottled up action tendencies
– A classroom with an authoritarian teacher
• Strong inhibitory forces lower Tr but not T
• Release of inhibition releases “bottled up action
tendency”

8
 Inhibition and Delay of onset
The effect of "bottled up" action tendencies
Personality as rates of change in states
35 Threat
removed

30
• What is stable is how rapidly one changes
25
• Sociability as rate of becoming sociable
20
Action and Negaction

15
Negaction
• Anxiety as rate of change of becoming anxious
10
• Intelligence as rate of change in problem space
5 • Need achievement as rate of growth in approach
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
motivation when faced with achievement goals
-5
Resultant Action

-10
Time

Negaction Action Resultant Action

Personality as rates of change Revised Dynamics of Action

• Growth rates, decay rates, inhibitory strengths • Cues
• Growth of tendency when stimulated • Action Tendencies
– dTa = personality x situation • Actions
• Decay of Ta when ongoing • Cues elicit action Tendencies
– Adaptation rate? • Tendencies strengthen actions
• Strength of inhibitory processes • Actions reduce Tendencies
• Decision rule is mutual inhibition

Cues, Tendencies, Action

Cues, Tendencies, Action
Compatible actions

Cues Tendency Action Cues Tendency Action

Cues Tendency Action

9
 Cues, Tendencies, Action
Computer simulations as formal theory
Incompatible actions
Cues Tendency Action
Cues Tendency Action
Additional alternative formulations
• General recognition of two motivations, two
types of behaviors, two outcomes
• Achievement motivation and approach
• Avoidance Motivation and withdrawal
• Promotion focus and approach
• Prevention focus and withdrawal
• Joy of gain, pain of loss
Elliot and Thrash, 2002
• Theory as a system of differential equations
• Simulations in terms of difference equations
• Predictions are consequences of the model
and are not always obvious
• Computer simulations of the CTA model
– Dynamic variables
Attributions and cognition
• Information gained by success and failure
– Success on hard tasks => high ability
– Failure on easy tasks => low ability
• Stability of self estimates of ability
• Stability of estimates of task difficulty
• Tasks as ways of learning vs. ways of
performing
10