Effect of Implicit Attitude on Automation Use

pucefakeAI and Robotics

Nov 30, 2013 (3 years and 8 months ago)

88 views

Introduction

Method

Result

Conclusion

Reference & Acknowledgement

Key Laboratory of Brain and Cognitive Science,

Institute of Psychology, Chinese Academy of Sciences

Jiaoyan

Yang,
Feng

Du,
Mingzheng

Wu,
Xianghong

Sun

Effect of Implicit Attitude on Automation Use

Automation

is

designed

to

set

people

free

from

manual

work

(
Parasuraman
,

Sheridan,

&

Wickens
,

2000
)
.

Previous

studies

indicated

that

whether

one

followed

automation’s

advice

depended

on

the

amount

of

trust

he

or

she

had

on

automation

(
Lee

&

See,

2004
)
.

However,

in

many

circumstance,

participants

disused

automation

even

when

they

trusted

automation

more

than

their

own

(
Parasuraman

&

Riley,

1997
)
.

The

dissociation

between

trust

and

automation

use

occurred

possibly

because

trust

was

measured

via

self
-
report
.

Self

report

provided

an

explicit

measure

of

trust

which

sometimes

couldn’t

reflect

participants’

actual

attitude

(
Chartrand
,

2005
)
.

Greenwald(
1995
)

proposed

implicit

attitude

which

referred

to

the

unconscious

trace

of

past

experience
.

A

previous

study

explored

individuals’

implicit

attitude

for

robots

and

t h e

r e s u l t s

s h o we d

t h a t

p a r t i c i p a n t s

r e l a t e d

r o b o t

mo r e

wi t h

u n p l e a s a n t

a s s o c i a t i o n s

c o mp a r e d

wi t h

h u ma n s (
Ma c Do r ma n
,

Va s u d e v a n
,

&

Ho,

2009
)
.

Based

on

the

study

of

robots,

we

assumed

that

the

implicit

attitude

for

automation

to

be

negative

and

that

it

might

partially

contribute

to

disuse

of

automation
.

Furthermore,

we

would

like

to

explore

whether

the

effect

of

implicit

attitude

on

automation

use

was

moderated

by

time

pressure
.

Participants and design

Fifty
-
three college students (20 males and 33 females) of Zhejiang University participated for
course credit. They were randomly assigned to one of the two time limit conditions: time
limited (1.5s) and unlimited.

Procedure

Participants undertook an implicit attitude test, explicit attitude test and automation use test .

1. Implicit Attitude Test: Go /No go task



Each

time

one

stimuli

word

of

four

categories

appeared

on

the

screen

for

1200
ms

.

Participants

responded

to

two

categories(one

target

category

and

attribute

category)

with

one

key

and

withhold

response

to

the

other

two

categories

before

the

stimuli

word

disappeared
.

In

the

present

study,

automation

and

human

expert

were

the

target

categories

while

pleasant

and

unpleasant

were

the

attribute

categories
.



2. Explicit Attitude Test: signal detection task



Participants

had

to

judge

whether

there

was

a

target

letter

X

among

the

56

green

alphabet

characters
.

Each

time

participants

made

the

judgment,

automation

would

reveal

its

judgment

but

it

was

only

75
%

accurate

in

judging

the

presence

of

the

target

letter

X
.

After

the

signal

detection

task,

participants

took

an

explicit

attitude

questionnaire

3. Automation Use Test: signal detection task



The

task

for

automation

use

test

was

similar

to

explicit

attitude

test

.

However,

for

these

trials,

after

participants

made

their

initial

judgment

and

received

feedback

(suggestions)

from

automation,

they

had

to

decide

for

a

second

time

whether

the

target

letter

X

appeared
.

The

time

participants

had

to

make

the

second

judgment

was

either

limited

to

1
.
5
s

or

unlimited
.

Fig

2
.

Signal

detection

task

for

explicit

attitude

test

and

automation

use

test


Abstract

Automation

often

falls

into

disuse

though

previous

studies

showed

that

participants

explicitly

trusted

automation

more

than

their

own

ability
.

Present

study

aimed

to

investigate

participants’

implicit

attitude

toward

automation

with

GNAT

(GO/NO
-
GO

Association

Task)

and

its

effect

on

automation

use
.

The

results

showed

dissociation

between

their

explicit

attitude

and

implicit

attitude
.

Participants

explicitly

perceived

automation

as

more

reliable

and

trusted

it

more

than

their

own
.

However,

their

implicit

attitude

toward

automation

was

negative,

resulting

in

a

very

high

disuse

rate
.

Further

analysis

also

confirmed

that

implicit

attitude

played

an

important

role

in

automation

disuse
.

More

specifically,

participants

with

a

positive

implicit

attitude

were

more

likely

to

rely

on

automation

than

those

whose

implicit

attitude

was

negative

or

neutral
.



Implicit

Attitude

and

Explicit

Attitude


Participants’

implicit

attitude

toward

automation

was

negative(as

shown

in

table

1
)
.

They

associated

automation

more

strongly

with

negativity
.

On

the

other

hand,

the

explicit

attitude

measure

revealed

that

participants

trusted

automation

(
M
=
1
.
36
,

SD
=
1
.
38
)

more

than

themselves

(
M
=
0
.
71
,

SD
=
1
.
69
),

t
(
51
)=
2
.
523
,

p
=
0
.
015
.












Effect

of

Implicit

Attitude

on

Automation

Use



Multiple

analysis

(R
2
=
0
.
243
)

revealed

that

implicit

attitude

was

a

significant

predictor

for

automation

use

(β=
0
.
273
,

t(
47
)=
2
.
105
,

p=
0
.
041
)
.

Moreover,

the

predictive

effect

of

implicit

attitude

was

not

moderated

by

time

limit
.

The

interaction

between

implicit

attitude

and

time

limit

was

not

significant

(β=
-
0
.
028
,

t(
47
)=
-
0
.
218
,

p=
0
.
829
)
.

Nor

was

the

main

effect

of

time

limit

(β=
-
0
.
118
,

t(
47
)=
-
0
.
926
,

p=
0
.
359
)
.




On

the

other

hand,

although

participants’

overall

implicit

attitude

toward

automation

was

negative,

there

were

eleven

participants

whose

implicit

attitude

was

positive

while

nineteen

participants’

was

neutral
.

Thus

we

classified

participants

into

three

groups

(positive,

neutral

or

negative)

according

to

their

implicit

attitude
.

ANOVA

analysis

showed

that

the

three

groups

had

significant

difference

on

automation

use,

F(
2
,
49
)=
3
.
446
,

p=
0
.
040
.

Participants

with

positive

implicit

attitude

relied

more

on

automation

than

those

whose

implicit

attitude

was

negative

or

neutral

(see

figure

1
)
.











Participants’

overall

implicit

attitude

toward

automation

is

negative
.


Implicit

attitude

plays

an

important

role

in

automation

use
.

Participants

with

positive

implicit

attitude

are

more

likely

to

rely

on

automation
.

We

would

like

to

thank

Richard

Carciofo

for

proofreading
.


Greenwald,

A
.

G
.
,

&

Banaji
,

M
.

R
.

(
1995
)
.

Implicit

Social

Cognition
:

Attitudes,

Self
-
Esteem,

and

Stereotypes
.

Psychological

Review,

102
(
1
),

4
-
27
.

Lee,

J
.

D
.
,

&

See,

K
.

A
.

(
2004
)
.

Trust

in

automation
:

Designing

for

appropriate

reliance
.

Human

Factors
:

The

Journal

of

the

Human

Factors

and

Ergonomics

Society,

46
(
1
),

50
-
80
.

MacDorman
,

K
.
,

Vasudevan
,

S
.
,

&

Ho,

C
.
-
C
.

(
2009
)
.

Does

Japan

really

have

robot

mania?

Comparing

attitudes

by

implicit

and

explicit

measures
.

AI

&

Society,

23
(
4
),

485
-
510
.


Parasuraman
,

R
.
,

Sheridan,

T
.

B
.
,

&

Wickens
,

C
.

D
.

(
2000
)
.

A

Model

for

Types

and

Levels

of

Human

Interaction

with

Automation
.

IEEE

Transactions

on

Systems,

Man,

and

Cybernetics
-
Part

A
:

Systems

and

Humans

30
(
3
),

286
-
297
.



Time unlimited

Time limited

M

S.D.

M

S.D.

Participants’

reliability


0.68

0.11

0.67

0.11

Implicit

attitude


-
0.11

0.33

-
0.11

0.36

Explicit

attitude


0.63

1.70

0.67

2.03

Automation

use


0.34

0.18

0.29

0.22

Table 1.

Means and standard deviations of the five central variables as a
function of experimental condition

0.00
0.10
0.20
0.30
0.40
0.50
0.60
negative
neutral
positive
Automation use

Implicit attitude

Fig 3.
Automation use across the three implicit attitude groups

Fig

1
.

Example

for

Go/No
-
Go

task
.

Participants

were

required

to

respond

to

automation

and

negative

while

hold

response

to

human

expert

and

positive
.

说明


1
、本模板尺寸


91*122 cm

36
*
48 inch), ICP 2012
要求为海报尺
寸不超过
97
*
194cm
。本板式宽度不可改变,,如果内容较多,可
增加调整高度,但不可小于
130cm



2
、请保持版面颜色设置、表
头及版块标题字体颜色和格式。正文

Time New Roman

Arial
,按照会议要求,字号不小于
16PT



3
、请根据隶属的
实验室选择标注