HSE 422 Lecture 4:

Anticipation & Decision Making

Rob Gray
 More than meets the eye:
• Last time we saw that expert athletes are
often better at searching their environments
to pick out crucial visual information.
• But, often times it seems like great athletes
know what’s going to happen before these
visual cues are available.
– E.g., great hitters know where the ball is going
before it leaves the pitcher’s hand.
 More than meets the eye:
• How do they achieve these magic feats?
Just what do they know that we don’t?
 More than meets the eye:

• Visual search = detecting relevant

information once the action begins to
unfold (e.g., after the ball is thrown or hit).
Vs.
• Anticipation = predicting how the action
will unfold before it happens (e.g., judging
where a tennis serve will go before the ball
is struck).
Specific Research Questions:
• Do experts remember more about the
environment around them?
• Do experts make use of contextual
information in anticipating future actions?
• Do experts make better use of situational
probabilities when anticipating?
• Do skilled performers make faster and
more accurate decisions?
Memory for Sport-Specific Information:
• Memory involves two basic stages:
– Encoding – transferring information into
permanent storage, from short-term memory
(STM) to long-term memory (LTM).
– Retrieval – getting the information out of
storage so it can be used.
Encoding

STM LTM

Retrieval
Environment
Rehearsal
 Sport Memory Research:
• Basketball (Allard et al., 1980)
– Showed experts and non-players
photographs of game situations
for 4 sec.
– Subjects were asked to remember
locations of players in photograph.
– Two types of scenes:
• Structured – picture taken during
play.
• Unstructured – scene during a time-
out.
 Sport Memory Research:
• Basketball (Allard et al., 1980)
– Results: Experienced basketball players could
accurately remember 8/10 players for
structured scenes while non-players could
only remember 5. Evidence for chunking,
leading to better encoding.
– There was no difference for the
unstructured.
– Key point: show advantage in memory that is
very specific to their sport. Don’t just have a
generally better memory!
 Sport Memory Research:
• Is this superior memory performance the
result of just more experience seeing
these type of scenes OR do you actually
have to play to get this improvement?
• Williams & David (1995).
– Compared memory performance for skilled
soccer players and wheelchair-bound fans
who had watched many games live and on
T.V. but had never played.
 Sport Memory Research:
• Williams & David (1995).
– Results: skilled players showed better ability
to recall compared with avid fans.
– Suggests that the memory advantage is
something they have developed through
practice and is not simply a by-product of
experience with the sport.
 Sport Memory Research:
• Bottom Line: Skilled athletes have larger
knowledge base of sports specific
information due to superior memory
processes.
– But do they make use of this information?
 Anticipation: Advance Cues
• A major component of anticipating what
will happen is the use of advance cues –
visual information that allows you to
predict what will happen before the
action begins
– E.g., the pitcher’s body language in baseball.
– Often essential to successful performance
because the action happens so fast!
 Advance Cues: Body Language

Biting tongue
= fastball
Slower arm
motion
= changeup More white of ball visible
= fastball

PSY 449- Human Factors in Sport – 13

Rob Gray
 Advance Cues: Research
• One popular method that has
been used is:
– The Occlusion Paradigm: show
players a scene of an unfolding
action and block-out (occlude)
part of it.
• E.g., show films of tennis serves with
server’s lower-body blacked out.
• Basic logic: if occluding something
makes an expert’s performance
worse they must be using this
information as an advance cue. occluder
 Advance Cues: Research
• Occlusion can also been done in terms of
time.
– For example, film of a sequence of a
baseball pitch can be edited to show three
different time intervals: (i) 500-300 msec
before ball release, (ii) 300-0 msec before
release, (iii) 0-200 msec after release..etc.
– Can reveal at what point in time the
performer begins to pick up information.
 Advance Cues: Research
• Several studies using the timing occlusion
technique have revealed that expert
performers can use advance cues to
predict the action:
– E.g., Ice hockey goaltenders can judge which
direction a shot is going accurately when the
film is stopped 500 msec before the shot!
– Novices inability to do this may be related to
confidence i.e., they don’t trust their guesses
and would rather wait for more information.
 Advance Cues: Research
• Abernethy et al.
– Presented film clips of badminton shots.
– Selectively occluded racket, face, body, lower
body or background behind player.
– Result: expert players need to be able to see
the racket and arm to know where the shot is
going.
 Advance Cues: Research
• More recently, researchers have used
advance technologies such as shutter
glasses which allow information to be
occluded during the performance of a real
action.
 Advance Cues: Research
• Most research has shown that the
predominant source of advance
information is the subtle body movements
of the thrower or hitter.
• The position of other players also seems to
be important.
– E.g., in baseball, a hitter can use the position
of the infielders as an advance cue to the pitch.
 Use of Situational Probabilities:
• In many sports, particular events are more
likely to occur (i.e., are more probable) in
particular game situations.
– E.g., in baseball batting certain pitches have a
higher probability depending on the pitch count.
• “Certainly the pitch you anticipate when the count is 0
and 2 (a curve ball probably, if the pitcher has one) is
not the pitch you anticipate when the count is 2 and 0
(fast ball, almost without exception).” Ted Williams.
• Do expert performers use this information?
 Use of Situational Probabilities:
• Gray (2002). Virtual baseball batting.
– Errors in the timing of a swing suggest that hitters
expect certain pitches for different counts.
65
msec
ERROR msec

60 fastballs

changeups
TEMPORAL ERROR

55

50
MEAN TEMPORAL

45

40

35
MEAN

30

25
0_1 1_0 1_1 0_2 2_0 1_2 3_0

PITCH COUNT
PSY 449- Human Factors in Sport – 21
Rob Gray
Use of Situational Probabilities:
• Gray (2002). Virtual baseball batting.
– Hitters also use the frequency of events to
anticipate the upcoming pitch.
– If a pitcher is throwing a lot of the same pitch,
the hitter will come to expect the same pitch
next time. Get similar results with racquet
sports.
– Emphasizes why it is important to key your
opponent off balance.
 Use of Situational Probabilities:
• Gray (2002). Virtual baseball batting.
0.8 F, F, F, F
MEAN SPATIAL ERROR m

0.7 F, F, F, S

0.6 EXPT. 1

0.5

0.4

0.3

0.2

0.1

0
1 2 3 4 5 6

BATTER
PSY 449- Human Factors in Sport – 23
Rob Gray
Use of Situational Probabilities:
• This type of knowledge is also used on an
individual level.
– E.g., knowing that your opponent in basketball
‘goes right’ most of the time when starting
from the top of the key.
 Decision Making:
• It’s great that expert athletes have all of this
extra information available but they still need
to decide what to do with it!
– Several studies (on basketball, hockey & field
hockey) have shown that experts make more
accurate decisions when shown photographs of
different sports situations. They are also usually
faster in making decisions.
– Most likely they have developed an elaborate set
of {if x then y} type of rules that they can quickly
pull up.
 Decision Making: Brain Cramps
• But this isn’t always the case!
• There have been some very famous
bonehead decisions in sport.
– Ian Woosnam was once penalized 2 strokes in
the British Open for carrying 15 clubs instead of
14!
– We will talk about these failures in performance
in a few weeks.
 Brain Activity & Anticipation:
• It is possible to measure brain activity while
someone is performing a sporting skill.
– Uses Electroencephalography (EEG)
• measures all of the electrical activity in the brain, not
from a specific area.
• when a brain wave is measured for a particular event
(like the pitcher releasing the ball), it is called an event
related potential (ERP).
• if you collect enough measurements you can identify
peaks & troughs in the electrical activity.
 The P300:
• A wave of
positive
electrical activity
that occurs
roughly 300
msec after a
stimulus
appears.
 Brain Activity & Anticipation:
• The latency of the P300 (how long it takes
to occur; e.g. 290ms vs. 310ms) is related
to the time required to recognize and
classify the object.
• The amplitude of the P300 is related to
task difficulty.
– easy tasks have a very small amplitude.
 P300 Predictions:
• Because skilled athletes rely more on
advance cues, situational probabilities etc.
than on perceptual cues
– they should have shorter P300 latencies
• Because skills are easier for skilled
athletes
– they should have smaller P300 amplitudes
 P300 Results:
• Suggests that intermediate batters are less
efficient in their perceptual-decision making
processes due to limitations in their
attentional capacity.
– Will take more about this next time…
 “Sports Software”: A Summary
• In general, we have seen that expert
athletes:
1. Are faster and more accurate in recognizing
patterns of play.
2. Search the visual field more quickly and
economically.
3. Are superior at anticipating actions on the basis
of advance visual cues.
“Sports Software”: A Summary

4. Have superior knowledge of situational

probabilities and make better use of them.
– And research has also shown that they:
5. Possess deeper and more structured knowledge
of both facts (e.g., rules) and procedures (‘how
to’).
6. Are better at self-monitoring to make
corrections in their play.
 “Sports Software”: A Summary

• But how did they get this way?

• In future lectures we will explore what types
of practice and training our needed to
develop this specialized software.
– We will also learn a few tricks we can try
ourselves!