A Study of Cognitive Effort of Decision Makers with Different NC

under Framing

Chiung-Wen Hsu

and Chen-li Kuo

Information Management Department, Cheng-Shiu University,

840 Chengcing Rd., Niaosong Dist., Kaohsiung City 83347, Taiwan

Department of Information Management, Chang Gung University,

259 Wen-Hwa 1st Road, Kwei-Shan Tao-Yuan, Taiwan, 333, Taiwan

Keywords: Framing Effect, Need for Cognition, Cognitive Effort, Eye-tracking.

Abstract: The purpose of this study is to examine cognitive effort of decision makers with different need for cognition

(NC) while making decision under framing. Hundreds of empirical studies have demonstrated the framing

effect moderating by NC in various contexts. However, these studies often treated cognition as a black box

and focused on the outcomes rather than on the process by which decisions with different NC are made. In

order to explore cognitive process of decision makers with different NC under framed problems, our

research observes the cognitive effort of decision makers with different NC (High vs. Low NC) under

different framing (Positive vs. Negative framing) from the perspective of their information process. A

laboratory experiment of 65 subjects was conducted. Eye-tracking was applied to evaluate decision makers’

cognitive effort. The results indicate that all subjects are susceptible to framing effect, and NC doesn’t

moderate framing effect. Decision makers with high NC will spend more cognitive effort to framed

problems. In addition, decision makers with high NC, compared with those with low NC, will pay more

cognitive effort in negative frame, but not for positive framing. Finally, there is no significant relationship

between cognitive effort and framing effect. The results could compensate the shortage of past studies

related to framing effect and NC, which only focused on final choices. In addition, by using eye tracking,

we also unveil the track of information process before framing effect generated, which could benefit the

richness of research on framing effect and NC.

1 INTRODUCTION

The formulation of a decision stimulus or an event

influences how decision makers think and decide.

Framing effect, proposed by Kahneman and Tversky

(1979), refers to the phenomenon that by presenting

the same problem in terms of gains (positive frame)

or losses (negative frame) can systematically affect

the choice one makes. The logically equivalent

problems are framed in positive versus negative

ways, so called valence-based framing, may

systematically affect the decisions or actions

decision makers take. People are prone to select the

risk-averse option under positive frame and the risky

option under negative frame.

These framing phenomena have been widely

investigated in a variety of research fields

(Kuhberger, 1998). A meta-analysis by Kuhberger

(1998) concluded that framing was a reliable

phenomenon and further suggested other variables

such as risk characteristics, task characteristics, and

personal personality might moderate the framing

effects. Need for cognition (NC), a personal trait, is

a significant factor moderating framing effect.

Cohen et al., (1955) defined NC as “the tendency for

an individual to engage in and enjoy thinking” (p.

116). Previous studies showed that people with

higher NC are willing to conduct more cognitive

processing, while those with lower NC are not

motivated to do so when making decisions.

Specifically, people with higher NC examine the

information more completely and tend to ignore

influences from other information (Verplanken,

1993). Because of the different thinking style

derived from NC, previous studies suggested that

NC is an important factor moderating framing effect.

Many studies found that NC interferes with framing

effect (Chatterjee et al., 2000); (Simon et al., 2004);

(Smith and Levin, 1996); (Petty et al., 2008),

suggesting that people with higher NC have more

Hsu C. and Kuo C..

A Study of Cognitive Effort of Decision Makers with Different NC under Framing.

DOI: 10.5220/0004437300750082

In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 75-82

ISBN: 978-989-8565-61-7

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

thinking on framing problems and are not

susceptible to framing effect. On the other hand,

recent research found that framing effect exists no

matter decision makers with high or low NC

(LeBoeuf and Shafir, 2003); (Levin et al., 2002);

(Cárdaba et al., 2013); (Tonetto and Stein, 2010).

Those studies, whether in support of the notion

that NC moderate framing effect, assume that level

of cognitive effort increases as NC increases (Simon

et al., 2004); (Petty et al., 2008). Cognitive effort

refers to the total use of cognitive resources required

to complete the task (Payne et al., 1990). However,

little studies use physiological apparatuses to

measure cognitive effort in order to provide

objective evidence. In addition, scarce studies

discuss which information is paid more attention in

response to framed problems. Therefore, one

purpose of this study is to measure and compare the

difference of cognitive effort devoted to the framed

problems between high and low NC decision

makers. Based on previous studies which show that

more cognitive effort paid to framing problems

reduce framing effect, the relationship between

cognitive effort and relationship is investigated

further.

Individual difference, NC, is a factor about one’s

ability to determine how cognitive effort paid to

framed problems. However, other variables such as

motivation to process may also moderate NC on

cognitive effort. Motivation could be driven by

complex tasks, relevant topics or untrusted messages

(See, Petty, and Evans, 2009). Overall, the

characteristics of task (i.e. level of complexity,

blatancy of message) are important factors

moderating NC on cognitive effort and effect of

persuasion (Petty et al., 2008). NC is more salient

depending on task conditions. For example, Levin et

al., (2000) set laptop shopping environment, and

asks respondents to choose the notebook they want

to buy. Each respondent has to use inclusion and

exclusion strategy to decide which product he/she

wants. The results revealed that in the inclusion

condition where subjects showed greater narrowing

of options (i.e. cognitive effort required), high NC

subjects processed information in a more focused

manner with greater depth and breadth than did low

NC subjects, and the quality of their selections

tended to be higher. The characteristics of task may

be important moderators for influence of NC on

cognitive effort, and worthy to explore further.

Recent breakthroughs in physiological

apparatuses such as functional magnetic resonance

imaging (fMRI) have helped disclose the underlying

cognitive activities during the process of decision

making. Another physiological apparatuses, eye

tracking, has been used to measure cognitive effort

in the studies related to decision making (Kuo et al.,

2009); (Huang and Kuo, 2012). Eye-tracking has

been used in a variety of research fields, ranging

from reading processing (Rayner,1998), marketing

(Wang and Day, 2007), and decision making (Kuo et

al., 2009); (Huang and Kuo, 2012). Studies which

track eye movements have proven that eye

movement is a sufficient and valid reflection of the

decision process (Kuo et al., 2009); (Huang and

Kuo, 2012). Researchers have confirmed the eye–

mind assumption: eye movements are directly

related to the underlying cognitive process (Rayner,

1998).

Thus, the present study utilized eye-tracking to

measure high and low-NC decision makers’

cognitive effort under framing. Further, the

moderator of framing (i.e. positive and negative

framing) on cognitive effect is investigated and the

relationship between cognitive effort and framing

effects is discussed. The result will explore the black

box into which decision makers’ cognitive effort

could be observed further under different framing.

2 LITERATURE REVIEW

AND HYPOTHESIS

2.1 NC and Cognitive Effort

Studies tackling the issue of whether NC interferes

with framing effect (Chatterjee et al., 2000); (Simon

et al., 2004); (Smith and Levin, 1996), suggested

that people with high NC analyse framed problems

more thoroughly. Most research hypothesizes that

different kinds of NC lead to difference cognitive

effort,. Scholars believed that people with low NC,

compared with high NC, are less motivated to

process information and have different information

processing and interpretation. People with high NC

are likely to process information in a careful,

elaborate fashion, paying less attention to peripheral

or superficial cues. They understand positive and

negative frame with the same equivalent logic, and

may not have cognitive bias by framing problems.

Verplanken (1993) showed the relationship between

individuals’ NC and cognitive effort measured by

variability of search across alternatives and pattern

of search. It was found that people with lower NC

applied less cognition effort to information

processing, and vice versa. Fisk and Neuberg’s

(1990) theory of impression formation demonstrated

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

the same hypothesis that people with high NC are

more motivated to analyse information, and the

analysis process turns from category to piecemeal

model. Thus, the present study hypothesized that

people with higher NC are more willing to put

cognition effort than people with low NC when they

are faced with framed problems.

H1: People with high NC pay more cognitive effort

than those with low NC under framing.

H1a: People with high NC pay more cognitive effort,

measured by number of fixations, than those with

low NC under framing.

H1b: People with high NC pay more cognitive effort,

measured by fixations duration, than those with

low NC under framing.

In addition to decision makers’ cognitive ability

influencing cognitive effort paid in the process of

decision, contextual factor is an important factor

motivating information process while making

decision. Prior NC research showed that different

tasks are of interest to those varying in NC.

Specifically, high NC individuals are more

motivated by complex tasks rather than simple (i.e.

high cognitive resources required) (See, Petty, &

Evans, 2009). Similarly, researchers claimed that

NC is internal traits, and it will only salient when

driven in ways such as instructing participants to

enter the information processing or motivating them

with complex tasks (Levin et al., 2000); (McElroy

and Seta, 2003). Both individual and situational

factors affecting the extent of thinking have been

identified in the persuasion process (See, Petty and

Evans, 2009).Levin et al., (2000) show that the NC

does indeed predict differences in information

search in decision making, and such differences are

more pronounced in situations where more cognitive

effort is needed. For example, Levin et al., (2000)

show that high NC subjects exhibited more effort

than low NC subjects only in the inclusion condition

which requires the most effortful thought. Thus, task

condition is an important factor for motivating

decision makers’ cognitive effort. Similarly, Simon

et al., (2004) found that the framing effect was

moderated by the combination of NC and the depth

of processing. NC is more salient by asking decision

makers to justify framed problems. Providing

reasons for making choices of framed problems

facilitates people with high NC, but not for those

with low NC, to think deeply and eliminate framing

effect.

In line with the arguments above, we assumed

that the influence of NC on cognitive effort will be

moderated by task characteristics. More complex

tasks will motivate decision makers’ NC to process

tasks. Studies have showed that negative framing is

more complex than positive framing, and cost

decision makers more cognitive effort to resolve

problems (Gonzalez et al., 2005). Therefore, we

hypothesize that valence of framing will moderate

the influence of decision makers’ NC on cognitive

effort. Specifically, people with high NC will pay

more cognitive effort under negative framing, but

not for positive framing, than those with low NC.

People with high NC will exert piecemeal-based

information processing, showing that more number

and duration of fixations are devoted to negative

framing. Hypothesis 2 is provided.

H2: People with high NC will pay more cognitive

effort under negative framing, but not for positive

framing, than those with low NC.

H2a: People with high NC will have more number of

fixations under negative framing, but not for positive

framing, than those with low NC.

H2b: People with high NC will have more fixation

duration under negative framing, but not for positive

framing, than those with low NC.

2.2 Cognitive Effort and Framing

Effect

Previous studies have shown that the framing effect

would be eliminated when more effort expended to

framed problems. Researchers claimed that the more

cognitive effort paid to framed problems by decision

makers, the more likely that framing effects can be

attenuated (Smith and Levin, 1996); (Chatterjee et

al., 2000); (Simon, 2004). For example, McElroy

and Seta (2003) argued individuals engaging a

decision task with an analytic processing style are

especially insensitive to the influence of framing

effect. Smith et al., (2004) also suggested that deep

thought manipulated by asking participants

justifying their choices would not be susceptible to

framing effect. Decision makes with high NC and

deep thought pay more attention on the relevant

attributes of the options and therefore eliminate

framing effect.

Cognitive effort paid to framed problems could

be derived from decision makers’ traits or be

manipulated by external forces. For the decision

makers’ traits, some studies revealed that people

with high NC can better reduce framing effect than

those with low NC (Smith and Levin, 1996);

(Chatterjee et al., 2000); (Simon, 2004). For the

external force, numerous studies ask subjects to

justify choices they made in order to increase

AStudyofCognitiveEffortofDecisionMakerswithDifferentNCunderFraming

thoughts given to framed problems. The deep

thoughts result more cognitive effort to process

information, and the framing effect weakens

(Takemura, 1994). In conclude, when decision

makers pay more cognitive effort to framed

problems, they may think more about the problem

and pay more attention on the relevant aspects of the

problems, consider the opposite information, such as

thinking about the death rate when only the survival

rate is presented. Thus, in line with previous studies,

hypothesis 3 is proposed. When decision makers pay

more cognitive effort on information processing

under framed problems, they can enter piecemeal-

based information processing, adopt more analytical

way to deal with framed problems and attenuate the

influence of framing.

H3: There is negative relationship between

cognitive effort and framing effect. The greater

cognitive effort is paid, measured by number of

fixations, the more likely it is that the framing effect

is reduced.

3 METHOD

3.1 Participants

A total of 65 college students with age ranging from

18-22 years were recruited as subjects from a

university in Taiwan. 15 participants were excluded

due to incorrect calibration and 50 participants are

valid. Cash reward were given for participation.

3.2 Stimulus Materials

Four risky-choice problems were employed in the

current research, including two in the life domain

and two in the monetary domain. All problems are

adopted from previous studies (Kahneman and

Tversky, 1979); (Kuhberger, 1995), and some

modifications are made in order to fit local contexts.

The problem description was validated by several

experts.

Each problem in this study had both a certain and

a risky alternative. The order in which the two kinds

of alternatives appeared in a problem was counter-

balanced. For each problem, two versions were

generated: the positively framed version and the

negatively framed version.

3.3 Measurement

(1) Cognitive Effort. Cognitive effort was measured

in the study by using EyeLink II with a sampling

rate of 1000Hz for tracking and recording subjects'

eye movements. Kahneman (1973) suggested that

cognitive effort be measured in terms of intensity

and time, both of which can be captured by means of

eye-fixation and eye-movement time. Thus, two

measurements of cognitive effort were employed in

this study: the average fixation per word and the

fixation duration per word.

(2) Need for Cognition. Need for cognition was

measured with the short form of the Need for

Cognition Scale (Cacioppo et al., 1984). The scale

consists of 18 items such as “I prefer my life to be

filled with puzzles that I must solve” that are rated

using 5-point Likert scales.

3.4 Design and Procedure

An incomplete within-subject design was employed

in our experiment. Subjects received both positive

and negative framing, but not for the same problem.

The 65 subjects were randomly divided into two

groups. Two positively framed problems and two

negatively framed ones were assigned to each group.

As four problems were used in this study, Group 1

was assigned the positively framed version of P1

and P2 as well as the negatively framed version of

P3 and P4. Conversely, Group 2 was assigned the

negatively framed version of P1 and P2 as well as

the positively framed version of P3 and P4. Four

framing problems were randomly presented to each

subject as stimuli.

Each participant was individually led to the

experimental room and asked to sit in front of the

experimental PC. The calibration and a subsequent

validation were treated. An experimental program

designed for the present study was then launched,

and the subject was told to read the instructions on

the screen and enter the choice via keyboard. Two

sample problems before the formal experiment are

presented in order to acquaint the subjects with the

experimental system and the eye tracker. After that,

four experimental problems were randomly

presented. At the beginning of each trial, the

problem and its two alternatives were displayed on

the screen. At the same time, the eye tracker started

to record subjects' eye movements. On average,

subjects took 15-30 seconds to finish a question, and

they were asked to enter their choice at any time

during a 30s response period. After finishing four

trials, participants are provided questionnaire about

their background and NC scale.

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

4 RESULT

The eye movement data from each participant were

inspected using a custom-made program to

determine whether or not the data were invalid due

to incorrect calibration (i.e., the fixation s were out

of screen positions). 15 participants' fixations were

invalid. The resulting data of 50 participants were

available for analysis. 26 subjects remained in

Group 1 and 24 in Group 2.

4.1 NC and Framing Effect

A SPSS-based χ2 test was conducted to examine

whether or not individuals made different choices

due to the difference in framing. Combing four

experimental framed problems, the proportion of

risk-seeking choices is 75% under negative framing

and 40% under positive framing (χ2 (1)=12.303,

p<0.05), respectively. The result indicates that

subjects in this study were more likely to choose the

risky alternative than the certain one in response to a

negatively framed problem. Besides, In order to test

the moderator of NC on framing effect, two-way

ANOVA is conducted to test the interaction. The

result shows that there was no significance of

interaction between framing and NC on framing

effect (F=1.70, p=0.21).

4.2 Hypothesis 1: NC and Cognitive

Effort

The coefficient alpha for the NC scale in the study is

0.79, showing good reliability. In order to test the

difference of cognitive effort paid by decision

makers with different NC, we divided participants

into high and low NC by using a median score, 3.45.

Participants’ NC scores higher than 3.45 are

categorized into high need for cognition. On the

other hand, scores lower than 3.45 are categorized

into low need for cognition. The result of t test,

shown as Table 1, indicates that there is difference

of NC scores between high and low NC group. T test

was used to analyse the differences between

cognitive efforts paid by decision makers with high

and low NC under framing and result are shown as

table 2.

The results indicate that subjects with high NC

also expended more frequency-based cognitive

effort, average fixation, to process framed problem

than those with low NC (M=1.62 per word vs.

M=1.39 ; t=2.584; p<.05). Similarly, subjects with

high NC in this study expended more cognitive

effort, measured by fixation duration, to process the

framed problem than those with low NC (M= 423.99

vs. M=374.82; t=2.143; p<.05). Therefore, both

Hypotheses 1a and 1b are supported. Collectively,

these results indicate that subjects with high NC in

this study exerted more cognitive effort for

processing information in framed problems than

those with low NC. Decision makers’ cognitive

effort paid for framed problems depends on their

NC.

Table 1: NC score between high and low NC group.

High NC Low NC t value P

NC Scores 3.84 3.22 6.109 <.05

Table 2: NC and Cognitive effort.

Cognitive

effort

High Low t value P

Average

fixation

1.62 1.39 2.584 <.05

fixation

duration (ms)

423.99 374.82 2.143 <.05

4.3 Hypothesis 2: Moderator of

Framing on Cognitive Effort

In order to test whether valence of framing interfere

with relationship between NC and cognitive effort.

Cognitive effort paid by decision makers with high

and low NC in positive and negative frame was

analysed separately. T test was conducted and

results are shown in Table 3 and 4.

Table 3: Positive frame: NC and cognitive effort.

Cognitive

effort

High Low t value P

Average

fixation

1.55 1.50 0.241 >0.1

fixation

duration (ms)

410.42 407.60 0.004 >0.1

Table 4: Negative frame: NC and cognitive effort.

Cognitive

effort

High Low t value P

Average

fixation

1.69 1.28 2.363 <.05

fixation

duration (ms)

437.56 342.04 4.283 <.05

Test result shows in positive frame, the average

fixation of high NC and that of low NC do not have

AStudyofCognitiveEffortofDecisionMakerswithDifferentNCunderFraming

distinct difference (M=1.55 vs. M=1.50; t=0.241,

p=0.812). Their fixation duration does not have

distinct difference either (M=410.42 vs. M=407.60;

t=0.004, p=0.951). However, in negative frame high

NC respondents’ and low NC respondents’ average

fixation have distinguished difference (t=2.363,

p<.05). High NC responds’ average fixation are

higher than that of low need for cognition

respondents (M=1.69 vs. M=1.28). Fixation duration

also has distinct difference (t=4.283, p<.05). High

need for cognition responds’ fixation duration is

longer than that of low need for cognition

respondents (M=437.56 vs. M=342.04). The H2a

and H2b are supported evidence by the test.

Respondents with high NC and low NC do not have

obvious difference of cognitive effort devoted to

positive frame, only in negative frame. Overall, the

information processing was more pronounced for

high-NC participants when they are faced with

negative framing.

4.4 Hypothesis 3: Cognitive Effort

and Framing Effect

χ2 was used to test the relationship between

cognitive effort and framing effect. First we split

cognitive effort into high and low cognitive effort by

a median of fixations for each framed problem.

Next, we count numbers of framing effect for each

problems framed as positive or negative. Framing

effect will be counted 1, if participants choose

certain option in positive frame or risky option in

negative frame, otherwise, they will be counted 0.

Then, two factors, cognitive effort and framing

effect, are tested by χ2. Cognitive effort and the

framing effect in the four problems were analysed

separately to test H3. Cognitive effort allocated to

positive and negative frame and framing effect are

not evident correlation. For the disease problem in

positive frame, the results of chi-square test is not

significant (χ2(1)= 0.11, p=0.74), neither in negative

frame(χ2(1)= 0.01, p=0.91). The second question

related to cancer in positive frame, the results of chi-

square test is not significant (χ2(1)= 0.12, p=0.72);

in negative frame they are also not significant

(χ2(1)=0.24, p=0.62). Third question related to

investment in positive frame, the results of chi-

square test is not significant (χ2(1)=2.40, p=0.12); in

negative frame they are also not significant

(χ2(1)=0.24, p=0.62). Fourth question related to

factory in positive frame, the results of chi-square

test is not significant (χ2(1)=0.26, p=0.61); in

negative frame they are also not significant

(χ2(1)=0.02, p=0.89). The above results show that

no matter which framed problem they are facing,

respondents’ cognitive effort and the reduction of

framing effect do not have distinguished correlation.

H3 is not supported.

5 DISCUSSION

Different from past studies, the main contribution of

this study is to trace cognitive effort allocated to

framing by decision makers with high or low NC.

From the observation of eye movement, the depth

and breadth of information process under framing

could be described specially and black box of

cognition could be explored further. In the present

study, H1 and H2 are supported and H3 not,

producing two important topics to be discussed.

1. NC and Cognitive effort. According to H1, we

found that decision makers with high NC will spend

more cognitive effort to process framed problems,

which is consistent with previous studies (Chatterjee

et al., 2000); (Simon et al., 2004); (Smith and Levin,

1996). From the distributions of fixations for framed

problems (see Figure 1), we found that participants

have more fixations on the two options than the

question. In addition, fixations on numerical

information such as probability and outcomes for

each option are more than other information no

matter decision makers with high or low NC.

Figure 1: Distributions of fixations for one framed

problem. Top panel: positive frame; bottom: negative.

Further, referred to H2, the valence of framing

will interact with need for cognition to determine the

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

extent of information processing. People who were

relatively high in their NC were more motivated to

process a message labelled as complex rather than

simple. Specially, people with high NC, compared

with those with low NC, will pay move cognitive

effort in negative frame. McElroy and Seta (2003)

suggest that some of the variability in decision

makers’ style of information process can be

accounted for by considering individual difference

and contextual factors. Specially, cognitive effort

only occurs when individuals are both willing (have

sufficient motivation) and able (have sufficient

motivation) to perform the task at hand. Therefore,

in situation where cognitive ability is not

constrained, motivational factors, such as the

complexity for task, emerge as the determining

cause for effort allocation. That is, as the complexity

of task increases, so too does the amount of effort

that an individual is willing to expend on the

decision task. The results in the study are also

consistent with findings provided by Smith et al.,

(2004) and Levin et al., (2000), which demonstrated

that task characteristics affect decision makers’ NC

more salient or not. The study by Levin et al. (2000)

show that decision makers with high NC will use

broader and deeper ways to find information than

those with low NC for inclusion strategy, not for

exclusion strategy. Accordingly, if task is easy,

decision makers’ NC may not motivated and

decision makers with different NC may not express

different thinking style. However, decision makers

with high NC may pay more cognitive effort to

process information when task is more complicated.

From the result, we know that NC may be salient or

not by the task conditions.

2. Cognitive Effort and Framing Effect. From H3,

we found that there is not definitely relationship

between cognitive effort and framing effect, which

is inconsistent with previous studies. A lot of

literature have explored that deep thought can reduce

framing effect. Deep thought will influence decision

makers’ cognitive styles, and will make them turn

their thinking style from holistic thinking style to

analytic thinking (McElroy and Seta, 2003); (Smith

et al., 2004). Hence, we infer that higher cognitive

effort can reflect stronger deep thought, and

eliminate framing effect.

Nevertheless, from our empirical data can notice

that high or low cognitive effort may not influence

framing effect. Petty and DeMarree, et al. (2008)

found that whether analytic thought derived from

high NC eliminate framing effect depending on

framing blatancy. As NC increases, the magnitude of

framing effects increases with a subtle prime but

decreases with a blatant prime. Besides, Smith et al.,

(2004) provide an explanation for not moderator of

NC on framing effect. They claim that except for

NC, math skill is an important factor influencing

framing effect. A decision maker with analytic

thinking style but not with math ability may not find

relevant attributes for attenuating framing effect.

Based on the argument by Smith et al., (2004), we

infer that the reason for not supporting H3 is that

when decision maker is dealing with framing

problem, even though they are searching information

more thoroughly, and have much information

attention, they don’t find relevant attributes to

eliminate framing effect if they don’t have enough

math skill.

6 CONCLUSIONS

Hundreds of empirical studies have demonstrated

the framing effect moderating by NC in many

different contexts (Chatterjee et al., 2000); (Simon et

al., 2004); (Smith and Levin, 1996); (LeBoeuf and

Shafir, 2003); (Levin et al., 2002), and explain

framing effect moderating by decision makers’ NC

by using cognitive information-processing

principles. However, no empirical evidence is

presented as the cognitive effort. Researchers

performing these studies often have treated

cognition as a black box by focusing on the

outcomes rather than on the process in which

decisions are made by decision makers with

different NC. As a result, the process of how people

with different NC pay cognitive effort to problems

framed with gain and loss has gone largely

unaddressed. Our findings offer an evidence for

cognitive effort paid by decision makers with

different NC in response to positively and negatively

framed problems. The results could compensate the

shortage of past studies related to framing effect and

NC, which only focused on final choices rather than

cognitive effort of decision makers under framing.

In addition, by using eye tracking, we also unveil the

track of information process before framing effect

generated, which could benefit the richness of

research on framing effect and NC.

REFERENCES

Cacioppo, J. T.; Petty, R. E.; Feinstein, J. A.; Jarvis, W. B.

G., (1996). Dispositional differences in cognitive

motivation: The life and times of individuals varying

in need for cognition. Psychological Bulletin, 119(2),

AStudyofCognitiveEffortofDecisionMakerswithDifferentNCunderFraming

197-253.

Cacioppo, J. T., Petty, R. E., & Kao, C. F., (1984). The

efficient assessment of need for cognition. Journal of

Personality and Assessment, 48, 306-307.

Cárdaba, M. A. M., Briñol, P., Horcajo, J., & Petty, R. E.,

(2013). The effect of Need for Cognition on the

stability of prejudiced attitudes toward South

American immigrants. Psicothema, 25(1), 73-78.

Chatterjee, S., Heath, T. B., Milberg, S. J., & France, K.

R., (2000). The differential processing of price in

gains and losses: the effects of frame and need for

cognition. Journal of Behavioral Decision Making,

13(1), 61-75.

Gonzalez, C., Dana, J., Koshino, H., and Just. M., (2005)

The framing effect and risky decisions: examining

cognitive functions with fMRI, Journal of Economic

Psychology, 26(1), 1–20.

Huang, Y.F., Kuo, F. Y., (2012). How impulsivity affects

consumer decision-making in e-commerce. Electronic

Commerce Research and Applications,11(6), 582-590.

Kahneman, D. (1973). Attention and Effort. New

Jersey:Prentice-Hall.

Kahneman, D., & Tversky, A., (1979). Prospect theory:

An analysis of decision under risk. Econometrica,

47(2), 263–291.

Kuhberger, A., (1995) The framing of decisions: a new

look at old problems, Organizational Behavior and

Human Decision Processes, 62 (2), 230–240.

Kuhberger, A., (1998) The influence of framing on risky

decisions: a meta-analysis, Organizational Behavior

and Human Decision Processes, 75 (1) 23–55.

Kuo, F. Y., Hsu, C. W., & Day, R. F. (2009). An

exploratory study of cognitive effort involved in

decision under Framing - an application of the eye-

tracking technology. Decision Support Systems, 48(1),

81-91.

LeBoeuf, R. A., & Shafir, E. (2003). Deep thoughts and

shallow frames: on the susceptibility to framing

effects. Journal of Behavioral Decision Making, 16,

77-92.

Levin, I. P., Gaeth, G. J., Schreiber, J., & Lauriola, M.

(2002). A new look at framing effects: distribution of

effect sizes, individual differences, and independence

of types of effects. Organizational Behavior and

Human Decision Processes, 88, 411-429.

Levin, I. P., Huneke, M. E., & Jasper, J. D., (2000).

Information processing at successive stages of

decision making: need for cognition and inclusion-

exclusion effects. Organizational Behavior and

Human Decision Processes, 82(2), 171-193.

McElroy, T., & Seta, J. J., (2003). Framing effect :a

analytic-holistic perspective. Journal of Experimental

Social Psychology, 39, 610-617.

Payne, J. W., Johnson, E. J., Bettman, J. R., and Coupey

E., (1990) Understanding contingent choice: a

computer simulation approach, systems, man and

cybernetics, IEEE transactions on systems, Man and

Cybernetics, 20(2), 296–309.

Petty, R. E., DeMarree, K. G., Briñol, P., Horcajo, J., &

Strathman, A. J. (2008). Need for cognition can

magnify or attenuate priming effects in social

judgment. Personality and Social Psychology Bulletin,

34(7), 900-912.

Rayner, K., (1998). Eye movements in reading and

information processing: 20 years of research.

Psychological Bulletin, 124(3), 372-422.

See, Y. H. M., Petty, R. E., & Evans, L. M. (2009). The

impact of perceived message complexity and need for

cognition on information processing and attitudes.

Journal of Research in Personality, 43(5), 880-889.

Simon, A. F., Fagley, N. S., & Halleran, J. G. (2004).

Decision framing: moderating effects of individual

differences and cognitive processing. Journal of

Behavioral Decision Making, 17, 77-93.

Smith, S. M., & Levin, I. P. (1996). Need for cognition

and choice framing effect. Journal of Behavioral

Decision Making, 9, 283-290.

Takemura, K., (1994). Influence of elaboration on the

framing of decision. The Journal of Psychology,

128(1), 33-39.

Tonetto, L. M., & Stein, L. M., (2010). Moderating Effects

of Consumer Involvement and the Need for Cognition

on Goal Framing. Revista Interamericana de

Psicologia/Interamerican Journal of Psychology,

44(2), 256-262..

Verplanken, B., (1993). Need for cognition and external

information search response to time pressure during

decision making. Journal of Research in Personality,

27, 238-252.

Wang, J.-C., & Day, R.-F., (2007). The Effects of

Attention Inertia on Advertisements on the WWW.

Computers in Human Behavior, 23(3), 1390-1407.

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems