bs_bs_banner

Geriatr Gerontol Int 2015; 15: 10791086

ORIGINAL ARTICLE: BEHAVIORAL

AND SOCIAL SCIENCES

Attention to negative emotion is related to longitudinal

social network change: The moderating effect of
interdependent self-construal
Tianyuan Li,1 Helene H Fung,2 Derek M Isaacowitz3 and Frieder R Lang4
1
Department of Psychological Studies and Center for Psychosocial Health and Aging, Hong Kong Institute of Education, 2Department of
Psychology, Chinese University of Hong Kong, Hong Kong, China; 3Department of Psychology, Northeastern University, Boston,
Massachusetts, USA; and 4Institute of Psychogerontology, Friedrich-Alexander-University of Erlangen-Nuremberg, Erlangen, Germany

Aim: Many previous studies have investigated older adults attentional preference toward different emotions.
Interdependent self-construal is identified to be an important moderator of this phenomenon. However, despite the
important social functions of emotions, the social consequence of older adults emotional preferences in attention
have not yet been examined. The current study tested how older adults attentional preferences assessed in the
laboratory influenced changes in their real-life social network, and how interdependent self-construal moderated this
effect.
Methods: A total of 45 older adults aged 6084 years participated in an eye-tracking session that measured their
attentional preference to emotional faces versus neutral faces. After that, participants completed the Self-Construal
Scale. Participants social network was then assessed by the Social Convoy Questionnaire twice over a 2-year period.
Results: Interdependent self-construal significantly moderated the effect of attention to angry and sad faces on older
adults real-life social network changes. For older adults with a higher level of interdependent self-construal, more
attention toward negative emotions was related to longitudinal decreases in the number of their emotionally close
social partners.
Conclusions: The present study shows the important role of attentional preferences in older adults social network
maintenance. It identified a real-life macro level social outcome of a micro level laboratory phenomenon, which can
be an important direction for future research. Geriatr Gerontol Int 2015; 15: 10791086.

Keywords: attention, interdependent self-construal, negative emotion, older adults, social network.

Introduction might have a significant impact on their social network.4

Several studies have investigated age differences in
attention to emotional stimuli, with older adults found
to attend more to positive than negative stimuli.1,2 Inter-
dependent self-construal is found to moderate this
effect, at least in collectivistic cultures.3 Given the
important role of emotions in the social context, older
adults attention toward different emotional stimuli
Accepted for publication 22 August 2014.
Correspondence: Professor Helene H Fung PhD, Department of
Psychology, The Chinese University of Hong Kong, Room 328
Sino Building, Chung Chi College, Shatin, New Territories,
Hong Kong, China. Email: hhlfung@psy.cuhk.edu.hk
However, while older adults selective cognitive process-
ing of emotional stimuli has been widely studied, its link
to real-life socioemotional outcomes is not well under-
stood.5 Although previous studies have investigated
how older adults managed their social network, no
study, to our knowledge, has tested the association
between older adults attention and their social network
characteristics.6,7 The current study attempted to fill in
this gap.
According to socioemotional selectivity theory, older
adults tend to allocate more of their cognitive resources
to process emotionally meaningful stimuli.8 One impli-
cation is that they show a bias toward positive informa-
tion, and away from negative information in their
attention, memory and other cognitive processing.1,9,10
However, a meta-analysis shows that the differences

2014 Japan Geriatrics Society doi: 10.1111/ggi.12401 | 1079

 T Li et al.
between older adults positivity preference and negativ-
ity preference in terms of memory or attention were not
significant.11
Why is older adults preference to positive over nega-
tive information identified in some samples, but absent
in a meta-analysis aggregating different samples? One
possible explanation is that there might be individual
and cultural differences in defining emotionally mean-
ingful information. Using eye-tracking techniques,
older Chinese adults were found to look away from both
happy and sad faces.3 Interdependent self-construal was
further found to be a moderator of older adults emo-
tional preferences in cognitive processing.12 Interdepen-
dent (versus independent) self-construal was proposed
by Markus and Kitayama to reflect how people construe
themselves and their relationships.13 People with inter-
dependent self-construal perceive themselves to be
highly connected with their social partners, and con-
struct their social relationships as part of their self iden-
tity. Harmonious relationships are one of their major
pursuits in life. Although older adults with lower inter-
dependent self-construal showed both positivity
enhancement effect (i.e. processing positive stimuli
better than neutral stimuli) and negativity reduction
effect (i.e. processing neutral stimuli better than nega-
tive stimuli), older adults with higher interdependent
self-construal only showed the positivity enhancement
effect, but not the negativity reduction effect.12 In other
words, older adults with higher interdependent self-
construal do not selectively avoid processing negative
stimuli.
However, although older adults preferences in cog-
nitive processing of emotional stimuli have been studied
in various contexts, the effects of such phenomenon on
older adults daily life have not been investigated.1,12 In
searching for the downstream real-life effects of the
emotional preferences, one area of particular interest is
older adults social network, because emotions play a
critical role in interpersonal relationships. Attention to
and perception of others emotion not only affect inter-
personal interaction,14 but also contribute to coordina-
tion within social groups, and might enhance the
adaptiveness of the groups in the evolutionary process.15
Older adults emotional preferences in attention could
have an important impact on their social network man-
agement. In the current literature, laboratory studies
usually focus on phenomena on the micro level. The
present study made a pioneer step to test whether a
micro level variable identified in the laboratory (i.e.
attentional preference) was related to a macro level
outcome (i.e. social network change).
Previous studies about older adults social network
distinguish emotionally close social partners (ECSP)
from peripheral social partners (PSP). ECSP are the
closest social partners whom the individual cannot
imagine life without; whereas PSP are important, but
1080 |
less close social partners who might help primarily in
achieving instrumental goals.16 Consistent with
socioemotional selectivity theory,8 adults in Western
societies were found to reduce their social network size
with age by deliberately dropping PSP while keeping
ECSP.6 Again, interdependent self-construal is found
to moderate this phenomenon. For participants with
higher interdependent self-construal, older age was
related to a larger number of ECSP and a stable number
of PSP.17 Furthermore, for people with higher interde-
pendent self-construal, age was not related to decrease
in the number of PSP across a 2-year interval.18 These
findings suggest that ECSP and PSP might both be
important to interdependent older adults, because
interpersonal relationships are indispensable compo-
nents of their self-identity.13
As older adults understanding about emotions and
social partners are both related to interdependent self-
construal, the effects of their emotional preference on
their social network change might also be moderated by
self-construal. In the absence of prior research on this
issue, and based on previous findings on eye gaze and its
impact, two possible hypotheses can be generated.
On the one hand, gaze toward anger or happy faces
was found to have real-time mood regulation effect for
older adults with high executive functioning.19 Looking
toward happy faces and away from angry faces helped
these older adults sustain their mood during a labora-
tory session. Applying similar rationale to the social
context, more attention toward positive emotions might
indicate to the older adults that their relationships are
going well; whereas more attention toward negative
emotions might highlight the negative experience in
relationships, and elicit a cycle of escalating negative
interactions between older adults and social partners. As
older adults tend to avoid conflicts in relationships, after
being reminded of their vulnerability in social relation-
ships, older adults might choose to shrink their social
networks in order to minimize potential conflicts.20 As
individuals with interdependent self-construal value
social relationships more, they might be more sensitive
to the ups and downs of their relationships.21 So, the
effects of selective attention to positive or negative emo-
tions are expected to be more salient for older adults
with high interdependent self-construal. Furthermore,
the social network change is expected to happen mainly
in regard to older adults ECSP. As people prioritize
emotionally meaningful goals in later adulthood, older
adults are expected to keep a closer monitor on their
emotionally close relationships, and be more responsive
to different emotional cues in these close relationships.8
To conclude, the first plausible hypothesis is that
attentional preference to negative emotions is maladap-
tive if older adults have an attentional preference to
negative emotions, the size of their social network, espe-
cially the number of their ECSP, might decrease over
2014 Japan Geriatrics Society
 Attention and social network
time. This would be particularly true for older adults
with high interdependent self-construal.
On the other hand, negative exchanges were found to
have certain beneficial effects for relationships in the
long term they were related to increased closeness of
social partners, and sustained relationship satisfaction
for highly committed people.22,23 Thus, negative emo-
tions might provide important social information that
alerts older adults about potential problems in their
social relationships. More attention toward negative
emotions could help older adults maintain their rela-
tionships, and even lead to expansion of their social
network. Similarly, such beneficial effects of selective
attention to negative emotions, if supported by the
results, are expected to be more salient for older adults
with high interdependent self-construal and more
apparent in ECSP. In other words, the second plausible
hypothesis is that attentional preference to negative
emotions is adaptive. If older adults have an attentional
preference to negative emotions, the size of their social
network, especially the number of their ECSP, might
increase. Again, this would be particularly true for older
adults with high interdependent self-construal.
These two alternative hypotheses were tested empiri-
cally in the current study.
Method
Participants
To test the aforementioned hypotheses, older adults
who completed an eye-tracking study about their
attentional preferences to emotional faces were invited
to participate in the current study.3 Younger partici-
pants of the earlier study were not invited to come back,
because the study did not find much variability in
younger adults attentional preferences. A total of 45 (38
women; Mage = 68.93 years, SDage = 5.86 years, range
6084 years) out of the 57 older adults from the original
study agreed to join the current study. Data about their
attentional preference collected from the eye-tracking
session were reanalyzed with the longitudinal data about
their social network change to address the present
research questions. All participants were Chinese.
Measures and procedure
After providing informed consent, participants took the
Snellen Visual Acuity Test and the PelliRobson Con-
trast Sensitivity Test, which measured their visual abili-
ties.24 An eye-tracking session was carried out next.
Attentional preferences to emotional faces
In each trial, participants were presented a synthetic
emotional face (i.e. angry, happy, sad, or fearful)
2014 Japan Geriatrics Society
together with a synthetic neutral face on a 15-inch com-
puter screen. The four types of emotional faces, the
position of emotional faces (i.e. left or right) and pre-
sentation time of each trial (1 s, 4 s or 8 s) were coun-
terbalanced in a total of 75 trials. Participants were
asked to watch the stimuli naturally in the way that they
would watch TV at home. Their fixation durations
toward the emotional face and the neutral face were
recorded by an Applied Science Laboratories Eye
Tracker (Model 504; Bedford, MA, USA). Attentional
preference to the emotional face in each trial was cal-
culated using the formula: (fixation to emotional
face fixation to neutral face) / (fixation to emotional
face + fixation to neutral face). Then, the indices calcu-
lated for all trials of the same emotion were averaged to
obtain the final index of attentional preference to each
emotion. More detailed information about the stimuli,
equipment and procedure of the eye-tracking session
was presented in a previous article.3
Social network at Time 1
Participants reported their social network characteristics
using the Social Convey Questionnaire.16,17 Three con-
centric circles were presented on an A4 page. The word
I was written in the middle of the circles. Participants
were asked to place their social partners into the three
concentric circles. Partners in the inner circles should
be the closest ones that one cannot imagine life
without. Partners in the middle circle should be close
to me, but not that close compared with people in the
inner circle. Partners in the outer circle should be not
that close to me, but still important. As in previous
studies, partners in the inner circle were counted as
ECSP, and partners in the middle and outer circle were
counted as PSP.17,18 The questionnaire has been widely
used as a valid measure of social network.2529 The
number of ECSP, but not PSP, is significantly related to
social embeddedness.7 The questionnaire has also been
shown to have high reliability in longitudinal studies.18,30
Interdependent self-construal
The Self Construal Scale was used to assess partici-
pants interdependent self-construal.17,31 The interde-
pendent subscale contains 14 seven-point Likert items
(Cronbachs = 0.63), such as I will sacrifice my self-
interest for the benefit of my group.
Cognitive ability and demographic information
Participants cognitive ability was measured by their
forward and backward digital spans. Participants also
reported their age, sex, religion and years of education.
Finally, participants rated their own health from poor (1)
to excellent (5). None of these background variables
| 1081
 T Li et al.

were significantly related to the longitudinal changes of Table 1 Descriptive statistics

social network, and thus were not included in further
analysis. Variables Mean SD
Age (years) 68.93 5.86
Years of education 6.91 3.81
Social network at Time 2
Subjective health 2.29 0.76
Participants were contacted 2 years later to report their Forward digital span 7.75 1.08
social network again. The measure of social network Backward digital span 3.48 1.05
was the same for Time 1. Longitudinal changes in social Snellen Visual Acuity Test 4.49 1.49
network were obtained by subtracting the number of PelliRobson Contrast 1.68 0.17
social partners at Time 1 from the number of social Sensitivity Test
partners at Time 2. We calculated the longitudinal Attentional preferences to
changes in the number of ECSP and PSP, respectively. emotional faces
Anger 0.02 0.18
Sadness 0.04 0.15
Results Fear 0.00 0.14
Happiness 0.04 0.17
Descriptive statistics of the sample and correlations Interdependent self-construal 5.80 0.36
between the variables are presented in Table 1 and ECSP at T1 7.00 3.46
Table 2, respectively. From Table 2, we can see that PSP at T1 9.53 5.20
attentional preferences toward the four emotions (i.e. ECSP at T2 6.24 3.37
anger, sadness, fear and happiness) did not significantly PSP at T2 10.29 5.01
correlate with older adults: (i) number of ECSP or PSP ECSP change 0.76 3.24
at either Time 1 or Time 2, r = 0.07 to 0.16, P > 0.301; PSP change 0.76 5.60
or (ii) change in the number of ECSP or PSP across the

2-year period, r = 0.13 to.08, P > 0.404. Scale for subjective health: 1 = poor, 2 = fair, 3 = good,
4 = very good, 5 = excellent. Forward digit span was
Multiple regression analyses were then carried out to measured in terms of how many single-digit numbers the
test the effects of attentional preference to emotional participants could remember and recite immediately after the
faces on older adults longitudinal social network experimenter. Backward digit span was measured in terms
changes, and whether such effects are moderated by of how many single-digit numbers the participants could
interdependent self-construal. In each regression remember and recite in reverse order immediately after the
experimenter. Results of Snellen Visual Acuity Test were
model, the dependent variable was either the change in transformed into numerical numbers according to the
the number of ECSP or the change in the number of following rule: 1 = 20/16, 2 = 20/20, 3 = 20/25,
PSP over the 2-year period. The predictors were 4 = 20/32, 5 = 20/40, 6 = 20/50, 7 = 20/63,
attentional preference to a specific emotion (standard- 8 = 20/80, 9 = 20/100, 10 = 20/125, 11 = 20/160,
ized), interdependent self-construal (standardized) and with larger numbers indicating poorer visual acuity. Results
from the PelliRobson Contrast Sensitivity Test showed the
the interaction between attentional preferences and log contrast sensitivity for the participants, with larger
interdependent self-construal. numbers showing better contrast sensitivity. ECSP,
emotionally close social partner; PSP, peripheral social
partner; T1, Time 1; T2, Time 2.
Attention to anger and social network change
The interaction between attention to anger and inter-
dependent self-construal was significant in predicting with lower interdependence (mean 2 SD), the pattern
the longitudinal change in number of ECSP, B = 1.26, was reversed, B = 2.22, SE = 1.01, = 0.70 and P =
SE = 0.53, = 0.39, P = 0.022, suggesting that interde- 0.034. Attention to anger was not related to the change
pendent self-construal significantly moderated how in older adults number of PSP, P > 0.91 for all
attention to anger influenced longitudinal changes in coefficients.
the number of older adults ECSP. The main effect of
attention to anger, B = 0.31, SE = 0.50, = 0.10 and
Attention to sadness and social network change
that of interdependent self-construal, B = 0.41, SE =
0.50 and = 0.13, were both non-significant. Simple The interaction between attention to sadness and inter-
slope analyses showed that for older adults with high dependent self-construal was also significant in predict-
interdependence (mean + 2 SD), more attention to ing the longitudinal change in number of ECSP,
anger was significantly related to more negative changes B = 1.36, SE = 0.64, = 0.35 and P = 0.039, suggest-
in the number of ECSP, B = 2.84, SE = 1.32, = 0.90 ing that interdependent self-construal also moderated
and P = 0.037 (see Fig. 1a). In contrast, for older adults the effect of attention to sadness on the longitudinal

1082 | 2014 Japan Geriatrics Society

 Table 2 Zero-correlations between variables
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

2014 Japan Geriatrics Society

1. Age
2. Year of education 0.03
3. Subjective health 0.04 0.12
4. Forward digital span 0.52** 0.09 0.04
5. Backward digital span 0.27 0.23 0.23 0.15
6. Snellen Visual Acuity 0.34* 0.14 0.11 0.37* 0.24
Test
7. PelliRobson Contrast 0.20 0.25 0.10 0.36* 0.24 0.44**
Sensitivity Test
Attentional preferences to
emotional faces
8. Anger 0.01 0.21 0.14 0.05 0.19 0.05 0.14
9. Sadness 0.10 0.16 0.42** 0.15 0.14 0.05 0.28 0.47**
10. Fear 0.05 0.31* 0.11 0.21 0.07 0.35* 0.16 0.19 0.01
11. Happiness 0.08 0.52** 0.05 0.12 0.01 0.07 0.16 0.34* 0.44** 0.45**
12. Interdependent 0.22 0.22 0.22 0.34* 0.16 0.10 0.04 0.11 0.18 0.08 0.21
self-construal
13. ECSP at T1 0.13 0.08 0.05 0.07 0.03 0.09 0.01 0.05 0.25
Attention and social network

0.16 0.30* 0.15

14. PSP at T1 0.27 0.01 0.04 0.28 0.20 0.09 0.05 0.04 0.09 0.07 0.16 0.16 0.01
15. ECSP at T2 0.11 0.13 0.00 0.13 0.18 0.07 0.07 0.06 0.02 0.09 0.13 0.24 0.55** 0.21
16. PSP at T2 0.04 0.07 0.09 0.09 0.01 0.03 0.08 0.03 0.09 0.07 0.03 0.18 0.50** 0.40** 0.14
17. ECSP change 0.26 0.03 0.08 0.08 0.13 0.09 0.15 0.03 0.07 0.08 0.08 0.02 0.50** 0.21 0.45** 0.38*
18. PSP change 0.21 0.05 0.05 0.18 0.19 0.06 0.02 0.01 0.00 0.13 0.12 0.01 0.43** 0.57** 0.07 0.52** 0.53**

*P < 0.05, **P < 0.01 (n = 45). Scale for subjective health: 1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent. Forward digit span was measured in terms of how many single-digit numbers the participants could

remember and recite immediately after the experimenter. Backward digit span was measured in terms of how many single-digit numbers the participants could remember and recite in reverse order immediately after the

experimenter. Results of Snellen Visual Acuity Test were transformed into numerical numbers according to the following rule: 1 = 20/16, 2 = 20/20, 3 = 20/25, 4 = 20/32, 5 = 20/40, 6 = 20/50, 7 = 20/63,

8 = 20/80, 9 = 20/100, 10 = 20/125, 11 = 20/160, with larger numbers indicating poorer visual acuity. Results from the PelliRobson Contrast Sensitivity Test showed the log contrast sensitivity for the participants,
with larger numbers showing better contrast sensitivity. ECSP, emotionally close social partner; PSP, peripheral social partner; T1, Time 1; T2, Time 2.

| 1083
 T Li et al.

(a) 4 INT: M-2SD happiness on the change in the number of either ECSP
INT: M or PSP, all P > 0.39, suggesting that attentional prefer-
2 INT: M+2SD ences to these two emotions were not associated with
older adults longitudinal social network changes.
0
EECSP change

-2 -1 0 1 2
-2
Discussion
-4
Previous studies have found that older adults emotional
-6 preferences in attention are moderated by their self-
construal. Older adults with high interdependent self-
-8 construal pay relatively more attention to negative
Aenon to anger (z-score) information.12 However, it has been an open question
whether such attentional preference is adaptive or not.
(b) 6
The current study took the first step toward investigat-
4 ing the social network consequences of older adults
attentional preferences toward emotions. Results sup-
2 ported our first hypothesis: for older adults with high
ECSP change

0
-2 -1 0 1 2 tive emotions was detrimental to their social network.
-2
-4
4
-6
-8
A
Aention
i to sadness ((z-score))
Figure 1 Interdependent self-construal moderated the
effect of attention to (a) anger and (b) sadness on the
longitudinal change of older adults emotionally close social
partners. ECSP, emotionally close social partner; INT,
interdependent self-construal; M, mean; SD, standard
deviation.
change of older adults number of ECSP. The main
effect of attention to sadness, B = 0.29, SE = 0.49 and
= 0.09, and that of interdependent self-construal,
B = 0.51, SE = 0.54 and = 0.16, were both non-
significant. Similar to the effect of attention to anger,
simple slope analyses found that for older adults with
higher interdependence (mean + 2 SD), more attention
to sadness was significantly related to more negative
changes in number of ECSP, B = 3.02, SE = 1.39,
= 0.93 and P = 0.036, whereas for older adults with
lower interdependence (mean 2 SD), the pattern was
reversed but not significant, B = 2.44, SE = 1.35 and
= 0.75, non-significant (see Fig. 1b). Again, attention
to sadness was not related to the change in older adults
number of PSP, P > 0.83 for all coefficients.
Attention to fear/happiness and social
network change
No significant effects were found in the regression
models estimating the effects of attention to fear or
interdependent self-construal, more attention to nega-
In particular, more attention to angry or sad faces
was related to more negative changes in the number
of ECSP over 2 years for older adults with higher
interdependence.
Anger and sadness are both emotions that show
others discontent toward the current status of the rela-
tionship. More attention to such emotions might lead
older adults to focus on the negative side of their rela-
tionships and elicit more conflicts. Furthermore,
harmony in social relationships is particularly important
for people with interdependent self-construal.13 Rela-
tionship harmony, in addition to self-esteem, is found
to be an important contributor to subjective well-being
for interdependent people.32 Thus, older adults with
high interdependent self-construal might be especially
sensitive and intolerant to discord with their social part-
ners, especially the emotionally close ones. They might
also be hurt more deeply when discord does occur.
Hence, when there is an attentional bias toward others
negative emotions, to avoid conflicts and to minimize
the risk of being hurt emotionally in relationships, inter-
dependent older adults might be more prone to drop
some troublesome close social partners from their social
networks. In contrast, for older adults with low interde-
pendent self-construal, as their self-worthiness might
be less dependent on the quality of their relationships,
they might be able to deal with relationship problems
more objectively and effectively. Thus, attentional pref-
erence to negative emotions could alert them of rela-
tionship problems, and help them maintain more close
social partners. As for why attentional preferences
toward emotions only affected older adults number of
ECSP, but not PSP, this might be attributable to the fact
that peripheral relationships were more instrumental
and less emotional, and thus were affected less by emo-
tional cues.

1084 | 2014 Japan Geriatrics Society

 Attention and social network
It has been found previously that older Chinese adults
had different attentional preference toward sad faces
versus angry faces. They looked away from sad faces,
but not from angry faces.3 Despite such difference, the
current study found that the impacts of attention to the
two emotions on older adults social network change
were very similar. Taken together, the findings jointly
suggest that older adults natural attentional preference
might not always be adaptive. In the current case,
although older Chinese adults did not avoid paying
attention to angry faces, such attention was maladap-
tive, as it was related to a decreased number of ECSP
over 2 years. Therefore, it is critical to empirically
examine the consequences of older adults attentional
preferences.
It is important to distinguish the current results from
the previous findings that people with high interdepen-
dent self-construal maintain or even increase their
social network with age.17,18 While the two previous
studies tested age differences in social network and how
self-construal moderated the effect, the current study
examined the consequence of older adults emotional
preference in attention on their social network changes.
Although older adults with high interdependent self-
construal have similar or even more expansive social
network sizes compared with their younger counter-
parts, the current results suggest that more attention to
negative emotions might be socially maladaptive for
these older adults, and could lead to shrinkage of their
number of ECSP.
It is also important to note that no main effects of
attentional preferences on social network changes were
identified in the current study. From Figure 1, we can
see that how attentional preferences affect social
network does not have a general pattern across indi-
viduals, but depends largely on self-construal. Thus, it
is important to consider self-construal in future social
network studies.
Another noteworthy point is that attention to happi-
ness and fear did not affect older adults social network
changes. Others happiness could signify that the status
quo of the relationship is good, and there is no need to
change; whereas fear is rarely present in older adults
social interactions.33 Therefore, attention to these two
emotions might have little effect on social network
changes. However, there could be cultural differences
regarding this phenomenon. Older adults in Western
societies, unlike their Eastern counterparts, show pref-
erence toward happy faces.1 Future studies are encour-
aged to test whether the current results can be
generalized to other cultures.
Finally, we acknowledge that the generalization of the
current finding might be limited by the relatively small
sample size. Furthermore, the current findings cannot
address the causal relationship between emotional pref-
erence in attention and social network change, as we did
2014 Japan Geriatrics Society
not have longitudinal data on attentional preferences.
Nevertheless, the current study is the first step to link
older adults attentional preferences with their social
network changes. Future studies can further test the
causal link from attention to social network change in a
larger sample, and examine whether the effects would
vary for older adults with different background charac-
teristics (e.g. residence arrangement, health status, etc.).
More generally, future studies are also encouraged to
follow the current direction to investigate real-life
macro level consequences of micro level psychological
phenomena identified in the laboratory.
Acknowledgment
Work on this study was partially funded by Hong Kong
Research Grant Council General Research Fund
#442813 to Helene Fung.
Disclosure statement
The authors declare no conflict of interest.
References
1 Isaacowitz DM, Wadlinger HA, Goren D, Wilson HR.
Selective preference in visual fixation away from negative
images in old age? An eye-tracking study. Psychol Aging
2006; 21: 4048.
2 Mather M, Carstensen LL. Aging and motivated cognition:
the positivity effect in attention and memory. Trends Cogn
Sci 2005; 9: 496502.
3 Fung HH, Isaacowitz DM, Lu AY, Wadlinger HA, Goren
D, Wilson HR. Age-related positivity enhancement is not
universal: older Chinese look away from positive stimuli.
Psychol Aging 2008; 23: 440446.
4 Keltner D, Haidt J. Social functions of emotions at four
levels of analysis. Cognition Emotion 1999; 13: 505521.
5 Isaacowitz DM, Blanchard-Fields F. Linking process and
outcome in the study of emotion and aging. Perspect Psychol
Sci 2012; 7: 317.
6 Lang FR. Regulation of social relationships in later adult-
hood. J Gerontol B Psychol Sci Soc Sci 2001; 56: P321P326.
7 Lang FR, Carstensen LL. Close emotional relationships in
late life: further support for proactive aging in the social
domain. Psychol Aging 1994; 9: 315324.
8 Carstensen LL. The influence of a sense of time on human
development. Science 2006; 312: 19131915.
9 Charles ST, Mather M, Carstensen LL. Aging and emo-
tional memory: the forgettable nature of negative images
for older adults. J Exp Psychol Gen 2003; 132: 310324.
10 Kisley MA, Wood S, Burrows CL. Looking at the sunny
side of life: age-related change in an event-related potential
measure of the negativity bias. Psychol Sci 2007; 18: 838
843.
11 Murphy NA, Isaacowitz DM. Preferences for emotional
information in older and younger adults: a meta-analysis of
memory and attention tasks. Psychol Aging 2008; 23: 263
286.
| 1085
 T Li et al.
12 Fung HH, Isaacowitz DM, Lu AY, Li T. Interdependent
self-construal moderates the age-related negativity reduc-
tion effect in memory and visual attention. Psychol Aging
2010; 25: 321329.
13 Markus HR, Kitayama S. Culture and the self: implications
for cognition, emotion, and motivation. Psychol Rev 1991;
98: 224253.
14 Van Kleef GA. How emotions regulate social life: the emo-
tions as social information (EASI) model. Curr Dir Psychol
Sci 2009; 18: 184188.
15 Spoor JR, Kelly JR. The evolutionary significance of affect
in groups: communication and group bonding. Group Proc
Intergr Relat 2004; 7: 398412.
16 Kahn RL, Antonucci TC. Convoys over the life course:
attachment, roles and social support. In: Baltes PB, Brim
OG, eds. Life-Span Development and Behavior. San Diego,
CA: Academic Press, 1980; 254283.
17 Yeung DY, Fung HH, Lang FR. Self-construal moderates
age differences in social network characteristics. Psychol
Aging 2008; 23: 222226.
18 Zhang X, Yeung DY, Fung HH, Lang FR. Changes in
peripheral social partners and loneliness over time: the
moderating role of interdependence. Psychol Aging 2011;
26: 823829.
19 Isaacowitz DM, Toner K, Neupert SD. Use of gaze for
real-time mood regulation: effects of age and attentional
functioning. Psychol Aging 2009; 24: 989994.
20 Charles ST, Piazza JR, Luong G, Almeida DM. Now
you see it, now you dont: age differences in affective
reactivity to social tensions. Psychol Aging 2009; 24: 645
653.
21 Kwan VSY, Bond MH, Singelis TM. Pancultural explana-
tions for life satisfaction: adding relationship harmony to
self-esteem. J Pers Soc Psychol 1997; 73: 10381051.
22 Fung HH, Yeung DY, Li K-K, Lang FR. Benefits of nega-
tive social exchanges for emotional closeness. J Gerontol B
Psychol Sci Soc Sci 2009; 64B: 612621.
1086 |
23 Li T, Fung HH. How negative interactions affect relation-
ship satisfaction: the paradoxical short-term and long-term
effects of commitment. Soc Psychol Pers Sci 2013; 4: 274
281.
24 Pelli DG, Robson JF, Wilkins AJ. The design of a new letter
chart for measuring contrast sensitivity. Clin Vis Sci 1988; 2:
187199.
25 Antonucci TC. Measuring social support networks: hier-
archical mapping technique. Generations 1986; 10: 1012.
26 Antonucci TC, Akiyama H. Social networks in adult life
and a preliminary examination of the convoy model. J
Gerontol 1987; 42: 519527.
27 Fung HH, Stoeber FS, Yeung DY-, Lang FR. Cultural
specificity of socioemotional selectivity: age differences in
social network composition among Germans and Hong
Kong Chinese. J Gerontol B Psychol Sci Soc Sci 2008; 63:
P156PP64.
28 Tkatch R, Artinian NT, Abrams J et al. Social network and
health outcomes among African American cardiac reha-
bilitation patients. Heart Lung 2011; 40: 193200.
29 Lang FR, Staudinger UM, Carstensen LL. Perspectives on
socioemotional selectivity in late life: how personality and
social context do (and do not) make a difference. J Gerontol
B Psychol Sci Soc Sci 1998; 53B: P21P30.
30 Lang FR. Endings and continuity of social relationships:
maximizing intrinsic benefits within personal networks
when feeling near to death. J Soc Pers Relat 2000; 17: 155
182.
31 Gudykunst WB, Lee CM. Assessing the validity of self
construal scales. Hum Commun Res 2003; 29: 253274.
32 Chen SX, Chan W, Bond MH, Stewart SM. The effects of
self-efficacy and relationship harmony on depression
across cultures: applying level-oriented and structure-
oriented analyses. J Cross Cult Psychol 2006; 37: 643658.
33 Charles ST, Carstensen LL. Unpleasant situations elicit
different emotional responses in younger and older adults.
Psychol Aging 2008; 23: 495504.
2014 Japan Geriatrics Society