Professional Documents
Culture Documents
To cite this article: Gisela Böhm & Hans-Rüdiger Pfister (2015) Attentional focus and anticipated
emotions in the face of future environmental risks: should I take the train or drive my car? /
Foco atencional y emociones anticipadas en vista de riesgos ambientales futuros: ¿debería ir en
tren o en mi coche?, Psyecology: Revista Bilingüe de Psicología Ambiental / Bilingual Journal of
Environmental Psychology, 6:1, 35-72
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Psyecology: Revista Bilingüe de Psicología Ambiental / Bilingual Journal of Environmental Psychology, 2015
Vol. 6, No. 1, 35–72, http://dx.doi.org/10.1080/21711976.2014.1002204
a high level of convenience and comfort; the ride itself may thus be anticipated to
be pleasant and enjoyable. Anticipating the negative environmental consequences
of car usage, in contrast, may be associated with negative emotions such as guilt
and regret concerning having contributed to pollution, resource depletion, and
climate change. The alternatives to using a car, for example biking or using public
transport, are all more effortful than taking the car. With respect to public transport
the decision maker may then anticipate an inconvenient ride on a crowded bus
when focusing on the behaviour, which contrasts with anticipating a clean and
pleasant future environment that results from using public transport when focus-
ing on the consequences.
The aim of the present study is to investigate whether environmental behaviour
is guided by anticipated emotions in a similar manner as it has been shown for
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
health risk behaviours. More specifically, we want to find out whether environ-
mentally harmful behaviour can be reduced or pro-environmental behaviour be
fostered by focusing people’s attention on the environmental consequences of
their behaviours and on the emotions that they anticipate to experience as reac-
tions to these consequences, compared to focusing their attention on the antici-
pated emotions that are associated with the behaviour. Adopting more pro-
environmental or lesser environmentally harmful behaviours can thus be regarded
as an expression of a problem-focused coping strategy that aims at reducing an
environmental problem and the threat emanating from it.
Following the design of Richard et al. (1996), car driving was chosen as the
environmentally harmful target behaviour. As the pro-environmental alternative,
we chose public transport, because we assumed that for routes that people drive
by car public transport would be the most relevant alternative for most people in
most situations. Note that it is not necessarily the case that a car ride is always
pleasant and a trip by public transport always unpleasant. A car ride can also be
strenuous and tiresome rather than joyful; likewise, a train ride can be comfortable
and relaxing rather than uncomfortable and effortful. Therefore, we manipulated
the focus of attention (on behaviour itself while driving one’s car, on behaviour
itself while traveling on a train, or on future environmental consequences) as well
as the valence of the experience (positive or negative). This resulted in six
between-subject conditions in which participants were induced to anticipate: (a)
a pleasant car ride; (b) an unpleasant car ride; (c) a pleasant trip by public
transport; (d) an unpleasant trip by public transport; (e) a pleasant clean future
environment due to a general reduction of car traffic; or (f) an unpleasant future
environment that is polluted from too much car traffic. We measured anticipated
emotions, self-reported car utilization and behavioural intentions with respect to
future transport mode choice. The same variables were measured at a follow-up
two weeks later.
Irrespective of the transportation mode, we expect that the positive valence
conditions trigger more intense positive and less intense negative anticipated
emotions than their negative valence counterparts.
Concerning the car driving conditions, we expected that anticipating the
negative future consequences of car traffic would induce more intense negative
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 39
and less intense positive emotions than anticipating a pleasant car ride.
Correspondingly, intentions to use the car should be lower when the negative
future conditions are anticipated compared to anticipating a pleasant car ride.
These two conditions correspond to the hedonic discrepancy of a pleasurable
behaviour that leads to regrettable consequences (Richard et al., 1996).
Anticipating an unpleasant car ride is expected to have a similar effect, compared
to anticipating a pleasant car ride, as anticipating the negative future consequences
of car driving. Again, anticipated negative emotions (e.g., anger about a traffic
jam) are expected to be more intense, positive anticipated emotions less intense,
and car utilization as well as car use intentions are expected to be lower than in the
pleasant car ride condition.
Our predictions for the public transport conditions mirror those for the car
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Method
Participants
Two hundred and thirty-nine undergraduate and graduate students at the
University of Education Ludwigsburg (Germany) participated in the study. All
majored in education or pedagogics. They received a monetary incentive of
€11.00 for their participation. Their age ranged from 19 to 39 years (M = 22.46,
SD = 3.41); 81.6% were female, which corresponds to the gender distribution at
the university.
environment.
Negative The protagonist travels The protagonist travels The protagonist
to work by car and to work by train and experiences an
experiences an experiences an unpleasant future
unpleasant and unpleasant and approx. 20 years from
stressful car ride (e.g., stressful train ride now during which car
search for parking, (e.g., delayed and traffic increased;
traffic jams). crowded train, heavy resulting in a polluted,
luggage). unhealthy and stressful
environment (e.g.,
smog, illnesses,
accidents).
Note: Each scenario existed in two versions with either a female or a male protagonist. Each
participant watched a protagonist of his or her own sex.
Session 2 was a follow-up that took place two weeks after Session 1. In
Session 2, participants showed up individually and filled in a questionnaire that
measured anticipated emotions, car utilization and behavioural intentions concern-
ing future travel mode choice. These measures were identical to the ones used in
Session 1. Session 2 lasted approximately 15 minutes. At the end of Session 2
participants were debriefed and thanked, and received their monetary incentive.
Measures
In the following, we will denote the first part of Session 1 that took place
immediately before the experimental manipulation as t0, the second part of
Session 1 after the manipulation as t1, and Session 2 as t2. The following
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
I always turn off the engine of my car’; (vii) ‘I always car pool’; (viii) ‘I know
specific departure times for buses or trains that depart from my home or my place
of work’; (ix) ‘For short distances (up to 2 km) I always do without my car’; (x)
‘The trunk of my car is always fully packed (reverse coded)’; (xi) ‘I am very
careful, only then to use a car when there is no alternative to it’.
These items were measured at t0 and t2. Item (ii) turned out to yield no
variance, presumably because participants did not have a roof rack; it was omitted
from further analyses.
Exploratory factor analyses of the remaining 10 items indicated that they did
not form a unidimensional scale. At both points in time, t0 and t1, the items
yielded a two-factor structure (maximum-likelihood FA with varimax rotation),
accounting for 32% (t0) and 33% (t2) of variance, respectively. Factor 1 repre-
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
sents environmentally friendly utilization of one’s car [items (i), (iv) to (vii), (ix)
to (xi), e.g., ‘For short distances I always do without my car’]. Factor 2 represents
familiarity with bus and train schedules [items (iii) and (viii), e.g., ‘I know
specific departure times for buses or trains …’]. As our intention was to measure
car utilization, we used only the items loading on Factor 1 to construct a Car
Utilization Scale from the estimated factor scores; the two items loading on Factor
2 were not used in further analyses. Cronbach’s alpha of the Car Utilization Scale
was .64 at t0 and .65 at t2. High scores on this scale indicate an environmentally
friendly style of car utilization.
Note that even though the items of this scale were adapted from an existing
scale, the exact collection and phrasing of these items are new and form an ad hoc
scale the validity of which is untested. However, we are not aware of any
established instruments measuring specifically the utilization of cars with respect
to environmental friendliness.
Anticipated emotions
Anticipated emotions were measured on ad hoc rating scales like the ones
commonly used in research on affective forecasting (cf., e.g., Gilbert & Wilson,
2007). Participants were instructed to anticipate their emotions with respect to
different events depending on their focus condition. Participants in the behaviour-
car condition were instructed to think of a particular route that they normally drive
by car, for example their way to work, and to imagine in as detailed a way as
possible the next time they would drive this route by car and how they would
experience this next car ride. For participants in the behaviour-public transport
condition the instructions were the same except that they were asked to think of a
route that they would take by public transport. Participants in the consequences
condition were instructed to imagine what the future would be like with respect to
traffic, air quality, environment and climate, and how they would experience this
future. All participants were asked which emotions the anticipated event evoked
in them.
Participants gave two ratings, one for the intensity with which they anticipated
positive emotions and one for the intensity of negative emotions.1 Both ratings
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 43
were given on a seven-point rating scale ranging from 1 (‘not at all’) to 7 (‘very
strongly’). Anticipated emotions were measured at t1 and at t2. At both points in
time each participant anticipated emotions with respect to the event that corre-
sponded to his or her experimental condition (behaviour-car, behaviour-public
transport or consequences). Anticipated emotions also served as a manipulation
check for the valence manipulation.
transport. The response scale ranged from -3 (‘definitely less often’) via 0
(‘equally often as before’) to +3 (‘definitely more often’).
Results
We will structure the results section according to the dependent variables. Each
dependent variable was analysed by means of a 3 (focus: behaviour-car vs.
behaviour-public transport vs. consequences) × 2 (valence: positive vs. negative)
× 2 (time: t0 vs. t2 or t1 vs. t2) analysis of variance with repeated measures across
time. Anticipated emotions and behavioural intentions were measured at t1 and t2
so that these are the two levels of the repeated measures factor Time for these
dependent variables. The Car Utilization Scale score was measured at t0 and t2 so
that for this dependent variable these two points of measurement form the levels
of the factor Time. All analyses were conducted with the mixed-effects function
lmer (Bates, Maechler, Bolker, & Walker, 2014) as part of the R statistical
software (R Core Team, 2013), using the Kenward-Roger approximation for
degrees of freedom (Kenward & Roger, 1997) due to unequal cell sizes, and the
multiple comparison procedures from Hothorn, Bretz, and Westfall (2008) for post
hoc contrasts. All analyses were conducted with and without frequency of car use
as a control variable. Including this control variable did not alter any of the
results; we therefore report the analyses without it.
Anticipated emotions
Intensity of anticipated positive emotions
The analysis of anticipated positive emotions yielded significant main effects for
focus and valence as well as significant two-way interactions for Focus × Valence
and Valence × Time.
Participants in the positive valence condition (M = 4.08, SD = 1.49) antici-
pated more intense positive emotions than participants in the negative valence
condition (M = 3.35, SD = 1.51), F(1, 230.62) = 33.81, p < .001, η2p = .09. This
effect primarily demonstrates that the valence manipulation was successful.
According to the main effect for focus, F(2, 230.53) = 68.62, p < .001,
η2p = .30, participants in the behaviour–car condition (M = 4.7, SD = 1.18)
anticipated more positive emotions for their next car ride than behaviour–public
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
transport participants did for their next trip by public transport (M = 3.61,
SD = 1.3), post hoc contrast z = –5.10, p < .001. Both behaviour conditions
elicited more intense anticipated positive emotions when contrasted with the
consequences condition (M = 2.72, SD = 1.48), z = 8.26, p < .001 (see
Figure 1). Thus, car is generally associated with more positive anticipated emo-
tions than public transport. Thinking about the future elicits the least positive
anticipated emotions.
The interaction between focus and valence is depicted in Figure 1,
F(2, 230.53) = 3.87, p = .022, η2p = .02. The valence conditions differ only in
the behaviour–public transport (z = –3.58, p = .005) and the consequences
conditions (z = –5.02, p < .001), but not in the behaviour–car condition.
Apparently, car driving is associated with positive emotions, which is not affected
by drawing attention to the unpleasant aspects of it.
The interaction between time and valence is depicted in Figure 2,
F(1, 221.74) = 11.53, p < .001, η2p = .01. Whereas the valence conditions differ
clearly at t1 (z = –5.25, p < .001), this difference is attenuated at t2, though still
7
Valence
6 positive
negative
Positive Emotion
1
Public Car Consequences
Focus
7
Valence
6 positive
negative
Positive Emotion 5
1
T1 T2
Time
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
significant (z = –2.58, p = .044). Thus, while the effect of the valence manipula-
tion on positive anticipated emotions weakened over time it still persisted after a
period of two weeks.
7
Valence
6 negative
Negative Emotion
positive
5
1
Public Car Consequences
Focus
7
Valence
6 negative
positive
Negative Emotion 5
1
T1 T2
Time
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Behavioural intentions
Intention to use the car in the future
With the intention to use a car in the future as the dependent variable, an analysis of
variance yielded a main effect of focus, F(2, 232.63) = 5.98, p = .003, η2p = .05,
and a significant interaction between focus and time, F(2, 226.79) = 3.90, p = .022,
η2p < .008. As depicted in Figure 5, at t1 participants in the consequences condition
(M = –.58, SD = .88) intended to reduce their car driving more than participants in
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 47
0.5
Focus
Public
• Car
Intention to use car Consequences
0.0
•
•
−0.5
−1.0
T1 T2
Time
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Figure 5. Intention to reduce car driving in the future as a function of time of measure-
ment and focus of attention; error bars represent 95% confidence intervals.
1.0
0.0
−0.5
Public Car Consequences
Focus
Figure 6. Intention to increase the use of public transport in the future as a function of focus
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
of attention and valence of experience; error bars represent 95% confidence intervals.
Mediation analyses
Our basic hypothesis assumes that future environmental risks trigger anticipated
emotions (e.g., anticipated regret) and that these emotions then promote pro-
environmental behaviours as a strategy to cope with the problem. Thus, antici-
pated emotions are assumed to mediate between the perception of an environ-
mental risk and environmental behaviour. In the context of the current study this
means that the experimental treatment shapes anticipated emotions, which then in
turn direct environmental behaviour. To test this assumption, we conducted a
mediation analysis. The most concise approach to testing a mediation assumption
would be to follow a structural equation modeling (SEM) approach. Our limited
sample size precluded a SEM analysis, though. We therefore followed the tradi-
tional mediation analysis approach proposed by Baron and Kenny (1986). Our full
3 × 2 experimental design is too complex for this approach; we therefore
simplified the design based on the results of the preceding analyses of variance.
We found that while the valence manipulation did affect anticipated emotions, the
effect of the focus manipulation was generally stronger. The focus effect was such
that the consequences condition differed from the behaviour conditions (car,
public transport), which did not differ themselves. Therefore, we considered
only focus as the independent variable in the mediation model and dichotomized
it into behaviour condition versus consequences condition (coded as 0 and 1,
respectively). Employing this dichotomous focus variable we tested the
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 49
the regression model then only focus remains significant. In sum, while both
anticipated emotions and behavioural intentions are affected by focus, we find no
evidence for the assumption that the relationship between focus and intentions is
mediated by emotions.
Discussion
We will first summarize the results. The anticipated emotions indicate that the
valence of experience was manipulated successfully. Participants in the positive
valence condition anticipated more intense positive and less intense negative
emotions than participants in the negative valence condition. Anticipated emo-
tions are also affected by the focus of attention. The consequence condition is
experienced more negatively, that is, more intense negative and less intense
positive emotions are anticipated, than in the two behaviour focus conditions.
Of those, a car ride is generally anticipated to be more positive and less negative
than a ride by public transport. Even though the effect of the valence manipulation
on anticipated emotions generally diminishes over time, positive anticipated
emotions are still more intense in the positive than in the negative condition
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
after two weeks. The induced negative anticipated emotions, in contrast, vanish
completely over time so that after two weeks the negative condition reaches the
level of the positive condition with respect to the intensity of anticipated negative
emotions.
While the valence manipulation did induce anticipated emotions, this effect
was not transferred to behavioural intentions. Both intention to use the car in the
future and intention to use public transport in the future were affected by the focus
of attention, but not by the valence of the experience. Particularly the consequence
condition differs from the two behaviour focus conditions. The intention to reduce
car driving as well as the intention to increase the use of public transport is
stronger in the consequence condition than in the two behaviour focus conditions,
which do not differ between each other. Note that in the negative consequence
condition a dire future environment was described that resulted from too much car
traffic, and in the positive environment condition a clean and healthy future
environment was described that resulted from a drastic reduction in car use.
Thus, both scenarios call for a reduction in car driving. Apparently, directing
the attention to future consequences influences intentions more strongly than a
luring or deterring description of the behaviour.
Concerning the intention to reduce car driving, focus of attention interacted
with time of measurement. The effect of focusing on future consequences was
unchanged after two weeks; focusing on a car ride in the behaviour-car condition
seems to have had a delayed effect on intentions in that participants in this
condition intended to reduce their car driving at t2 but not at t1. Similarly, general
car use behaviour as measured by the Car Utilization Scale was slightly more
environmentally friendly at t2 than at t1.
Imagining the future and its affective impact may influence our current
decisions (Gilbert & Wilson, 2007). Common knowledge implies that car driving
will deteriorate the global environment in the long run, whereas using public
transport will be beneficial for the environment. Our findings suggest that direct-
ing attentional focus on these future consequences in fact influences one’s antici-
pated emotions and intentions to use cars or public transport. The effects
identified in this experiment were fairly small, but so was the experimental
treatment consisting of only a three-minute slide show. Interestingly, a focus on
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 51
manipulation. This would have allowed us to analyse more specifically what the
habitual reactions of our participants to the various travel modes were and whether
our manipulation induced changes in these habitual patterns.
A large proportion of our sample was female. Women have been found to
experience emotions concerning behavioural outcomes more strongly than men
(Eriksson & Simpson, 2010). Also, women tend to have weaker car habits and to
be more willing to reduce car use (Matthies, Kuhn, & Klöckner, 2002). These
results may suggest that the effects that we found may be overestimated by our
predominantly female sample. However, we believe that our sample had very
positive attitudes towards cars, as suggested by the positive emotions that were
associated with a car ride and unaffected by the negative car scenario. Our
participants stem from a region that is strongly shaped by a major German car
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Note
1. Participants also rated the intensity of a number of concrete anticipated emotions
(e.g., sadness, fear, regret, hope, joy, happiness). These concrete emotions formed
two factors in a factor analysis, corresponding to positive and negative emotions,
respectively. These two factors yielded the same results as the two overall ratings for
positive and negative emotions and will therefore not be reported in this paper.
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 53
Método
Participantes
Participaron en el experimento 239 estudiantes de grado y de posgrado de la
Universidad de Educación de Ludwigsburg (Alemania), y todos ellos cursaban
estudios de educación o de pedagogía. Recibieron un incentivo monetario de
11.00€ por su participación. Sus edades oscilan entre 19 y 39 años (M = 22.46,
DT = 3.41); el 81.6% eran mujeres, lo que se corresponde con la distribución por
género de la universidad.
Diseño y procedimiento
Se seleccionaron dos opciones de transporte: conducir el coche propio como la
opción supuestamente perjudicial para el medio ambiente, y viajar en tren, es
decir, en transporte público, como la opción supuestamente pro-ambiental. Se
manipularon dos variables independientes, el foco de atención y la valencia de la
experiencia. El foco de atención se desglosaba en tres niveles: el comportamiento
de conducir un coche, el comportamiento de utilizar el transporte público y las
consecuencias medioambientales; la valencia de la experiencia se desglosaba en
dos niveles, de modo que era positiva o negativa, resultando en seis condiciones
inter-sujetos.
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 57
Mediciones
En adelante nos referiremos a la primera parte de la Sesión 1 que tuvo lugar
inmediatamente antes de las manipulaciones experimentales como t0, a la segunda
parte de la Sesión 1 después de las manipulaciones como t1 y a la Sesión 2 como
t2. Se midieron las siguientes variables dependientes:
montado’ (codificación inversa); (iii) ‘Tengo a mano los horarios de los trenes o
autobuses que pasan por mi domicilio o por mi lugar de trabajo’; (iv) ‘Hago
excursiones al campo con mi coche’ (codificación inversa); (v) ‘Pongo mucho
empeño en organizar mis trayectos en coche de modo que los viajes que hago solo
son los mínimos necesarios’; (vi) ‘Cuando tengo que esperar mucho (e.g., en
semáforos o cruces ferroviarios), siempre apago el motor del coche’; (vii)
‘Siempre comparto el coche’; (viii) ‘Me sé los horarios de los autobuses o trenes
que pasan por mi domicilio o lugar de trabajo’; (ix) ‘En las distancias cortas (hasta
2 km.) nunca utilizo el coche’; (x) ‘El maletero del coche siempre está lleno’
(codificación inversa); (xi) ‘Pongo mucho empeño en utilizar el coche solo
cuando no hay otra alternativa’.
Estos ítems se midieron tanto en t0 como en t2. El ítem (ii) no produjo
varianza, seguramente porque los participantes no disponían de portaequipajes
en el vehículo, por lo que se omitió en los análisis posteriores.
Los análisis factoriales exploratorios de los 10 ítems restantes indicaron que
estos no formaban una escala unidimensional. En ambos momentos, t0 y t1, los
ítems produjeron una estructura bi-factorial (AF de máxima verosimilitud con
rotación varimax) que explica unas varianzas de 32% (t0) y 33% (t2) respectiva-
mente. El Factor 1 representa un uso pro-ambiental del vehículo propio (ítems (i),
(iv) a (vii), (ix), e.g., ‘En las distancias cortas nunca utilizo el coche’). El Factor 2
representa familiaridad con los horarios de trenes y autobuses (ítems (iii) y (viii),
e.g., ‘Me sé los horarios de los trenes y autobuses …’). Puesto que nuestra
intención era medir el uso del coche, utilizamos solo los ítems con carga en
relación al Factor 1 para construir una Escala de uso del coche a partir de los
resultados estimados de los factores; los dos ítems con carga factorial 2 no se
utilizaron en los análisis posteriores. El coeficiente alfa de Cronbach de la Escala
de uso del coche era de .64 en t0 y de .65 en t2. Los altos valores de esta escala
indicaban un uso pro-ambiental del coche.
Hay que señalar que, aunque los ítems de esta escala habían sido adaptados a
partir de otra escala preexistente, la recopilación y expresión exacta de estos ítems
era nueva y dio lugar a una escala ad hoc cuya validez no se ha comprobado. Sin
embargo, no conocemos ninguna herramienta establecida que mida de forma
específica el uso del coche en relación con un comportamiento pro-ambiental.
60 G. Böhm and H.-R. Pfister
Emociones anticipadas
Las emociones anticipadas se midieron utilizando escalas ad hoc como las que
suelen utilizarse en la investigación sobre predicciones afectivas (cf., e.g., Gilbert
& Wilson, 2007). Se solicitó a los participantes que anticiparan sus emociones
respecto a las diferentes situaciones según su condición de foco. A los partici-
pantes con la condición comportamiento-coche se les pidió que pensaran en un
trayecto particular que solían hacer en coche —por ejemplo, su trayecto al trabajo
— y que imaginaran del modo más detallado posible la próxima ocasión que
hubieran de conducir ese trayecto y cómo se sentirían en ese próximo trayecto.
Para los participantes con la condición comportamiento-transporte público, las
instrucciones eran las mismas, solo que se les pidió que pensaran en un trayecto
que harían habitualmente en transporte público. A los participantes en la
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Resultados
Estructuraremos la sección de resultados de acuerdo con las variables depen-
dientes. Cada variable dependiente se analizó utilizando un análisis de varianza
3 (foco: comportamiento-coche frente a comportamiento-transporte público
frente a consecuencias) x 2 (valencia: positiva frente a negativa) x 2 (tiempo:
t0 frente a t2 o t1 frente a t2) con mediciones repetidas en el tiempo. Se
midieron las emociones anticipadas y las intenciones de comportamiento en t1
y t2 para que estos fueran los niveles de las mediciones repetidas del factor
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 61
Tiempo para estas variables dependientes. Los valores de la Escala de uso del
coche se midieron en t0 y t2 de modo que para esta variable dependiente estos
dos momentos de medición conforman los niveles del factor Tiempo. Todos los
análisis se llevaron a cabo con la función de efectos mixtos lmer (Bates,
Maechler, Bolker, & Walker, 2014) como parte del software estadístico R (R
Core Team, 2013), utilizando la aproximación Kenward-Roger de los grados de
libertad (Kenward & Roger, 1997) debido a los tamaños de celda desiguales, y
los procedimientos de comparación múltiple de Hothorn, Bretz, y Westfall
(2008) para los contrastes post-hoc. Todos los análisis se realizaron con y sin
frecuencia de uso del coche como variable de control. La inclusión de esta
variable de control no modificó ninguno de los resultados y, por lo tanto, se
presenta el informe de los análisis sin ella.
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Emociones anticipadas
Intensidad de las emociones anticipadas positivas
Los análisis de las emociones anticipadas positivas produjeron efectos principales
importantes para el foco y la valencia, así como interacciones bidireccionales
significativas entre Foco x Valencia y Valencia x Tiempo.
Los participantes en la condición de valencia positiva (M = 4.08, DT = 1.49)
anticipaban emociones positivas más intensas que los participantes en la
condición de valencia negativa (M = 3.35, DT = 1.51), F(1, 230.62) = 33.81,
p < .001, η2p = .09. Este efecto demuestra principalmente que la manipulación de
la valencia se realizó con éxito.
Según el efecto principal para el foco, F(2, 230.53) = 68.62, p < .001,
η2p = .30, los participantes en la condición comportamiento–coche (M = 4.7,
DT = 1.18) anticipaban emociones más positivas para su próximo viaje en
transporte público (M = 3.61, DT = 1.3), contraste post–hoc z = –5.10,
p < .001. Ambas condiciones de comportamiento produjeron emociones positivas
más intensas en comparación con la condición de consecuencias (M = 2.72,
DT = 1.48), z = 8.26, p < .001 (véase Figura 1). Así pues, el coche se asocia
generalmente con emociones anticipadas más positivas que el transporte público.
La consideración del futuro provoca las emociones anticipadas menos positivas.
La interacción entre foco y valencia se describe en la Figura 1,
F(2, 230.53) = 3.87, p = .022, η2p = .02. Las condiciones de valencia varían tan
62 G. Böhm and H.-R. Pfister
7
Valencia
6 Positiva
Emoción Positiva negativa
5
1
T. Público Coche Consecuencias
Foco
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Figura 1. Intensidad de las emociones anticipadas positivas como función del foco de
atención y la valencia de la experiencia; las barras de error representan intervalos de
confianza de 95%.
7
Valencia
6 Positiva
Emoción Positiva
negativa
5
1
T1 T2
Tiempo
Figura 2. Intensidad de las emociones anticipadas positivas como función del tiempo de
medición y la valencia de la experiencia; las barras de error representan intervalos de
confianza de 95%.
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 63
7
Valencia
6 negativa
Emoción Negativa
positiva
5
1
T. Público Coche Consecuencias
Foco
7
Valencia
6 negativa
Emoción Negativa
positiva
5
1
T1 T2
Tiempo
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Figura 4. Intensidad de las emociones anticipadas negativas como función del tiempo de
medición y la valencia de la experiencia; las barras de error representan intervalos de confianza
de 95%.
Intención de comportamiento
Intención de uso del coche en el futuro
Con la intención de uso del coche en el futuro como variable independiente, el
análisis de varianza produjo un efecto principal del foco F(2, 232.63) = 5.98,
p = .003, η2p = .05, y una interacción significativa entre el foco y el tiempo, F(2,
226.79) = 3.90, p = .022, η2p < .008. Como se muestra en la Figura 5, en t1 los
participantes en la condición de consecuencias (M = –0.58, DT = 0.88) declaran
su intención de reducir su uso del coche más que los participantes en las
condiciones de comportamiento (comportamiento–coche: M = –0.15, DT = 0.86;
0.5
Foco
Intención de uso del coche
T. Público
Coche
0.0
Consecuencias
−0.5 −1.0
T1 T2
Tiempo
Figura 5. Intención de reducir el uso del coche en el futuro como función del Tiempo de
medición y el foco de atención; las barras de error representan intervalos de confianza de 95%.
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 65
1.0
Valencia
negativa
positiva
0.5
0.0
−0.5
T.Público Coche Consecuencias
Foco
Análisis de Mediación
Nuestra hipótesis básica presupone que los riesgos medioambientales futuros provo-
can emociones anticipadas (e.g., arrepentimiento anticipado) y que estas emociones su
vez promueven conductas pro-ambientales como una estrategia de afrontamiento del
problema. De este modo, se presupone que las emociones anticipadas actúan como
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Discusión
En primer lugar resumiremos los resultados. Las emociones anticipadas indican que la
valencia de la experiencia fue manipulada con éxito. Los participantes en la condición
de valencia positiva anticipaban emociones positivas más intensas y emociones
negativas menos intensas que los participantes en la condición de valencia negativa.
Las emociones anticipadas también se ven afectadas por el foco de atención. La
condición de consecuencias se vive de un modo más negativo, es decir, los partici-
pantes anticipan emociones negativas más intensas y emociones positivas menos
intensas que en las dos condiciones centradas en el comportamiento. En estas
últimas, el trayecto en coche suele anticiparse de un modo más positivo y menos
negativo que el trayecto en transporte público. Aunque el efecto de la manipulación de
la valencia en las emociones anticipadas suele disminuir en el tiempo, tras dos
68 G. Böhm and H.-R. Pfister
coche como un mayor uso del transporte público se consideran maneras de evitar
consecuencias medioambientales negativas y de fomentar las positivas.
Hemos podido identificar un efecto en el comportamiento pro-ambiental similar
al que se ha documentado en el comportamiento relativo a la salud. Cuando el
individuo se centra en las consecuencias a largo plazo de su comportamiento
ambiental, en lugar de centrarse en aspectos a corto plazo del propio comporta-
miento, este anticipa estados emocionales negativos; presumiblemente, la amenaza
y el arrepentimiento son componentes importantes de estas emociones anticipadas.
También podríamos demostrar que centrando el foco de atención en las consecuen-
cias en lugar de en el comportamiento dirige las intenciones de comportamiento
hacia un comportamiento más pro-ambiental y menos perjudicial para el medio
ambiente. Estos resultados están en línea con los obtenidos por Joireman, van
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Lange, y van Vugt’s (2004), que demostraron que las preferencias de transporte
de la gente toman forma principalmente en base a su consideración futura. Así pues,
tanto a nivel afectivo como a nivel de comportamiento, observamos que las con-
secuencias perjudiciales para el medio ambiente inducen la amenaza y un compor-
tamiento de afrontamiento que tiene por objeto solucionar el problema
medioambiental. Sin embargo, no encontramos pruebas de que la respuesta emo-
cional negativa mediara en el efecto e influyera en el comportamiento.
Si bien el comportamiento relativo a la salud y el ambiental presentan muchas
similitudes, también difieren en algunos aspectos importantes. Posiblemente sean
las diferencias entre los dos ámbitos las causantes de que no se observara un papel
de mediador en las emociones anticipadas. Una diferencia importante es que los
comportamientos sanitarios de riesgo suelen producir riesgos personales. Es decir,
la persona que realiza un comportamiento poco saludable es también la misma
que sufre sus posibles consecuencias negativas. En cambio, las consecuencias de
los comportamientos ambientales son colectivas. Se ha sugerido que las elec-
ciones ambientales pueden conceptualizarse como un dilema social que opone los
intereses individuales a los colectivos (van Vugt, van Lange, & Meertens, 1996).
Asimismo, la causa de un problema medioambiental no suele ser una acción
individual, sino los efectos acumulados de acciones colectivas. Así pues, tanto
las causas como las consecuencias de los problemas medioambientales son
colectivas (Böhm & Pfister, 2008b; Pfister & Böhm, 2001). Por el contrario, la
elección del medio de transporte es una decisión individual. Posiblemente, cen-
trarse en las consecuencias a largo plazo hace surgir emociones colectivas tales
como la culpa y la amenaza colectivas, que pueden ser independientes de las
emociones personales que dictan el comportamiento ambiental individual.
Tanto las emociones como las intenciones parecen no verse afectadas por las
circunstancias agradables de los comportamientos. Una posible explicación sería que
los comportamientos cotidianos como la elección del medio de transporte son
altamente habituales y están dictados por supuestas restricciones (Verplanken,
Aarts, & van Knippenberg, 1997). Así pues, es posible que el enfoque en los aspectos
negativos o positivos de un trayecto en coche no tenga ningún efecto porque la gente
hace tiempo ya que se ha adaptado a las condiciones cambiantes de sus comporta-
mientos habituales. Podría habernos sido de ayuda obtener mediciones de base para
70 G. Böhm and H.-R. Pfister
del coche, como sugieren las emociones positivas asociadas al trayecto en coche, y
que no se veían afectadas por el escenario negativo sobre el mismo. Nuestros
participantes son originarios de una región que está muy condicionada por un
gran fabricante de coches alemán, y la universidad no está muy bien conectada al
transporte público, lo que apoya nuestra opinión. No disponemos de datos
empíricos que la sustenten, sin embargo. Así pues, existe una necesidad clara de
investigación adicional para analizar el papel que desempeñan la atención y la
previsión afectiva en una muestra más equilibrada, así como para investigar com-
portamientos ambientales menos habituales y menos restringidos por factores
circunstanciales tales como la disponibilidad de transporte público.
Nota
1. Los participantes también valoraron la intensidad de determinadas emociones anticipa-
das concretas (e.g., tristeza, temor, arrepentimiento, esperanza, alegría, felicidad). Estas
emociones específicas formaban dos factores en un análisis factorial, correspondientes a
las emociones positivas y negativas respectivamente. Estos dos factores produjeron los
mismos resultados que las dos valoraciones generales para las emociones positivas y
negativas y, por lo tanto, no se ha informado sobre ello en este estudio.
Acknowledgements / Agradecimientos
This research was supported by the German Research Council (grant BO 1480/1-3 to the
first author). We wish to thank Stefanie Rhein, Sonja Schlegel, Anja Vocilka, Susanne
Jaggy and Petra Baisch for their support in conducting the experiment. / Esta
investigación ha sido realizada con el apoyo del Consejo Alemán de Investigación
(German Research Council), a través de la beca BO 1480/1-3 concedida al primer
autor. Deseamos dar las gracias a Stefanie Rhein, Sonja Schlegel, Anja Vocilka,
Susanne Jaggy y Petra Baisch por su apoyo en la realización de este experimento.
References / Referencias
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173–1182. doi:10.1037/
0022-3514.51.6.1173
Attentional focus and anticipated emotions / Foco atencional y emociones anticipadas 71
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2014). lme4: Linear mixed-effects
models using Eigen and S4. R package version 1.0.6. http://CRAN.R-project.org/
package=lme4
Böhm, G., & Pfister, H.-R. (2000). Action tendencies and characteristics of environmental
risks. Acta Psychologica, 104, 317–337. doi:10.1016/S0001-6918(00)00035-4
Böhm, G., & Pfister, H.-R. (2008a). Anticipated and experienced emotions in environ-
mental risk perception. Judgment and Decision Making, 3, 73–86.
Böhm, G., & Pfister, H.-R. (2008b). Antinomies of environmental risk perception:
Cognitive structure and evaluation. In M.J. Casimir (Ed.), Culture and the changing
environment. Uncertainty, cognition, and risk management in cross-cultural perspec-
tive (pp. 61–77). Oxford: Berghahn.
Böhm, G., & Tanner, C. (2013). Environmental risk perception. In L. Steg, A. E. Van Den
Berg, & J. I. M. De Groot (Eds.), Environmental psychology: An introduction
(pp. 15–25). New York, NY: Wiley-Blackwell.
Eriksson, K., & Simpson, B. (2010). Emotional reactions to losing explain gender
Downloaded by [University of Veracruzana] at 16:40 30 July 2015
Skinner, E. A., Edge, K., Altman, J., & Sherwood, H. (2003). Searching for the structure
of coping: A review and critique of category systems for classifying ways of coping.
Psychological Bulletin, 129, 216–269. doi:10.1037/0033-2909.129.2.216
Van der Pligt, J. (2002). Cognition and affect in risk perception and risky decision-
making. In C. Von Hofsten & L. Bäckman (Eds.), Psychology at the turn of the
millennium, social, developmental, and clinical perspectives (Vol. 2, pp. 247–270).
Hove: Psychology Press.
van Vugt, M., van Lange, P. A. M., & Meertens, R. M. (1996). Commuting by car or
public transportation? A social dilemma analysis of travel mode judgments. European
Journal of Social Psychology, 26, 373–395. doi:10.1002/(sici)1099-0992(199605)
26:3<373::aid-ejsp760>3.3.co;2-t
Verplanken, B., Aarts, H., & van Knippenberg, A. (1997). Habit, information acquisition,
and the process of making travel mode choices. European Journal of Social
Psychology, 27, 539–560. doi:10.1002/(SICI)1099-0992(199709/10)27:5<539::aid-
ejsp831><539::AID-EJSP831>3.0.CO;2-A
Downloaded by [University of Veracruzana] at 16:40 30 July 2015