Final Report Summary - SKCDM (Self-Knowledge in Consumer Decision Making)
The main objective of the project was to use a phenomenon called choice blindness to study self-knowledge and preference formation, particularly in the domain of consumer choice and economic decision making. Choice blindness (CB) refers to the experimental finding that people sometimes fail to notice mismatches between their intended choice and the outcome they are presented with, while nevertheless offering introspectively derived reasons for why they chose this alternative rather than the other (Johansson et al., 2005; Johansson et al., 2006). An integral component of the research is also the development of novel research methods, as the use of techniques borrowed from close-up card magic.
In an article named 'Magic at the marketplace', published 2010 in the journal Cognition, we demonstrated the CB effect for the taste of jam and the smell of tea in an ecologically valid supermarket setting (Hall, Johansson, Tärning, Sikström & Deutgen, 2010). Moving further into consumer psychology, we have created a near-identical mock-up version of one of the world's largest retail shopping sites. Here, we allow participants to freely shop for different (hypothetical or real) goods. The products are displayed in the manner they are normally presented on the website, and offer the same type of interaction. At the check-out stage, the participants are asked to rate how influential each attribute have been for their decision to buy the chosen item. At this point, the price or other features of their purchase are manipulated. In a recently completed full scale study we found that very few participants discovered these manipulations, and this despite being asked very specific questions about why they preferred the item chosen (Johansson, Hall, Christoforidis, Kusev, Aldrovandi & Chater, in prep), thus establishing CB for consumer choice in a typical consumer setting. In addition, we find a pronounced preference effect, as the participants come to prefer the alternative they were led to believe they liked when faced with the same choice again. This effect also generalise, as the participants change their overall shopping strategy as a result of the manipulation, for example going from preferring low cost-low quality products to high price-high quality products in subsequent choices.
In Johansson, Hall and Chater (2011) we discuss this temporal dynamics of the choice blindness situation, i.e. what effect the manipulation has on the participant's memory of the event and attitudes towards the manipulated stimulus. This argument is further extended in Johansson, Hall and Gärdenfors (2011), in which we elaborate on the implications of 'preference change through choice' from an evolutionary perspective.
In addition, we have also conducted a series of experiments on CB and choices between safe and risky alternatives, finding a similar pattern. Not only do the participants fail to detect the changes made, they also adjust their preferences for risk in accordance with the manipulation, for example reversing their overall risk preference from risk seeking to risk aversion (Kusev, Johansson, Ayton, Hall, van Schaik & Chater, in prep). We have recently discussed these results in a position paper, spelling out the main theoretical implications of this work in relation to different theories of risk preference (Chater, Johansson & Hall, 2011), and we are now working on a formal model connecting these results into a more general theory of choice (Kusev, Tsaneva-Atanasova, van Schaik & Johansson, in prep).
Since the publication of Johansson et al. (2005) we have been repeatedly challenged to demonstrate that choice blindness extends to domains such as moral reasoning, where decisions are of greater subjective importance, and where deliberation and introspection are seen as crucial ingredients of the process (e.g. Moore & Haggard, 2006). In a recent paper (Hall, Johansson & Sandberg, under review) we introduce 'the magical questionnaire' (see http://www.lucs.lu.se/cbq/(se abrirá en una nueva ventana) for an illustrative video), a new methodology that allows us to surreptitiously alter participants' written responses in questionnaire surveys. In this study, we establish that even for moral dilemmas hotly debated in the current press (such as the criminal status of prostitution, or whether file sharing of copyrighted material should be legalised) participants may fail to notice when they are exposed to alterations of their choices just a few minutes after having stated their answers. Surprisingly, this can be seen even when participants express very strong opinions about the subject matter, resulting in consequent confabulation about outcomes that are the polar opposite of the original choice.
Taking this one step further, in the lead-up to the Swedish election, we applied a version of the magical questionnaire to the Swedish electorate (Hall, Johansson, Strandberg, Lind, Tärning & Pärnamets, in prep). This time, we used an 'election compass', which is a battery of questions capturing the differences in view between the two major political coalitions in Sweden (with a focus on traditional issues in the conservative-socialist divide, such as taxation and privatisation). The sum of the score on this set of question is meant to help people decide which coalition that best fits their own views; by manipulating enough of the answers in the battery we could therefore change their overall allegiance. The results showed that only 20 % of the manipulated answers were noticed by the participants, which allowed us to move the aggregate scores of 91 % of all participants across the partisan dividing line (i.e. only 9 % of the participants detected enough manipulations to adjust their summary score back to the original coalition profile). Thus, we managed to create a situation where a large majority of the participants accepted a preference reversal across the socialist-conservative divide. In addition, the participants often changed their final stated voting intention in line with the manipulations made, thus showing that the choice-induced preference change may be found even in something as subjectively important as stated voting intention (see http://www.lucs.lu.se/cbp/(se abrirá en una nueva ventana) for a video demonstration).
In a very recent study, performed in collaboration with researchers from Unilever, we have combined the new methodology with our previous insights from choice blindness in consumer choice (Johansson, Hall, Sandberg, Dijksterhuis & Chater, in preparation). Here we have adapted an industry standard evaluation questionnaire used in focus group valuation of new products. Again, most of the participants fail to detect changes made to their evaluations (e.g. which product they find more natural, healthy, tasty, overall appeal, etc), and they also show the same type of preference reversal seen in other domains.
In the future of self-control (Hall, Johansson & de Léon, in press), we also present a general theoretical framework for how to address problems of self-control (for example, how to quit smoking) by using external cognition to increase our self-understanding. This line of argument is anchored in our empirical studies on choice blindness, as they show how we are often unaware of our fundamental lack of self-knowledge, while at the same time this lack can be alleviated by the use of external sources of information.
The studies I have conducted during my time as a Marie Curie fellow have shown that the choice blindness methodology can make a valuable contribution to the study of self-knowledge in consumer decision making, both in and outside the lab, and that the choice blindness situation (i.e. accepting and arguing for a choice not made) induces a preference change in a variety of experimental contexts, including consumer choice, monetary gambles and political opinion. These results are important in a number of different theoretical contexts (behavioural economics, decision making under risk, polling and voting behaviour, etc), as well as for applied consumer choice and marketing research.
Contact:
Petter.Johansson@lucs.lu.se
Website: http://www.lucs.lu.se(se abrirá en una nueva ventana)
In an article named 'Magic at the marketplace', published 2010 in the journal Cognition, we demonstrated the CB effect for the taste of jam and the smell of tea in an ecologically valid supermarket setting (Hall, Johansson, Tärning, Sikström & Deutgen, 2010). Moving further into consumer psychology, we have created a near-identical mock-up version of one of the world's largest retail shopping sites. Here, we allow participants to freely shop for different (hypothetical or real) goods. The products are displayed in the manner they are normally presented on the website, and offer the same type of interaction. At the check-out stage, the participants are asked to rate how influential each attribute have been for their decision to buy the chosen item. At this point, the price or other features of their purchase are manipulated. In a recently completed full scale study we found that very few participants discovered these manipulations, and this despite being asked very specific questions about why they preferred the item chosen (Johansson, Hall, Christoforidis, Kusev, Aldrovandi & Chater, in prep), thus establishing CB for consumer choice in a typical consumer setting. In addition, we find a pronounced preference effect, as the participants come to prefer the alternative they were led to believe they liked when faced with the same choice again. This effect also generalise, as the participants change their overall shopping strategy as a result of the manipulation, for example going from preferring low cost-low quality products to high price-high quality products in subsequent choices.
In Johansson, Hall and Chater (2011) we discuss this temporal dynamics of the choice blindness situation, i.e. what effect the manipulation has on the participant's memory of the event and attitudes towards the manipulated stimulus. This argument is further extended in Johansson, Hall and Gärdenfors (2011), in which we elaborate on the implications of 'preference change through choice' from an evolutionary perspective.
In addition, we have also conducted a series of experiments on CB and choices between safe and risky alternatives, finding a similar pattern. Not only do the participants fail to detect the changes made, they also adjust their preferences for risk in accordance with the manipulation, for example reversing their overall risk preference from risk seeking to risk aversion (Kusev, Johansson, Ayton, Hall, van Schaik & Chater, in prep). We have recently discussed these results in a position paper, spelling out the main theoretical implications of this work in relation to different theories of risk preference (Chater, Johansson & Hall, 2011), and we are now working on a formal model connecting these results into a more general theory of choice (Kusev, Tsaneva-Atanasova, van Schaik & Johansson, in prep).
Since the publication of Johansson et al. (2005) we have been repeatedly challenged to demonstrate that choice blindness extends to domains such as moral reasoning, where decisions are of greater subjective importance, and where deliberation and introspection are seen as crucial ingredients of the process (e.g. Moore & Haggard, 2006). In a recent paper (Hall, Johansson & Sandberg, under review) we introduce 'the magical questionnaire' (see http://www.lucs.lu.se/cbq/(se abrirá en una nueva ventana) for an illustrative video), a new methodology that allows us to surreptitiously alter participants' written responses in questionnaire surveys. In this study, we establish that even for moral dilemmas hotly debated in the current press (such as the criminal status of prostitution, or whether file sharing of copyrighted material should be legalised) participants may fail to notice when they are exposed to alterations of their choices just a few minutes after having stated their answers. Surprisingly, this can be seen even when participants express very strong opinions about the subject matter, resulting in consequent confabulation about outcomes that are the polar opposite of the original choice.
Taking this one step further, in the lead-up to the Swedish election, we applied a version of the magical questionnaire to the Swedish electorate (Hall, Johansson, Strandberg, Lind, Tärning & Pärnamets, in prep). This time, we used an 'election compass', which is a battery of questions capturing the differences in view between the two major political coalitions in Sweden (with a focus on traditional issues in the conservative-socialist divide, such as taxation and privatisation). The sum of the score on this set of question is meant to help people decide which coalition that best fits their own views; by manipulating enough of the answers in the battery we could therefore change their overall allegiance. The results showed that only 20 % of the manipulated answers were noticed by the participants, which allowed us to move the aggregate scores of 91 % of all participants across the partisan dividing line (i.e. only 9 % of the participants detected enough manipulations to adjust their summary score back to the original coalition profile). Thus, we managed to create a situation where a large majority of the participants accepted a preference reversal across the socialist-conservative divide. In addition, the participants often changed their final stated voting intention in line with the manipulations made, thus showing that the choice-induced preference change may be found even in something as subjectively important as stated voting intention (see http://www.lucs.lu.se/cbp/(se abrirá en una nueva ventana) for a video demonstration).
In a very recent study, performed in collaboration with researchers from Unilever, we have combined the new methodology with our previous insights from choice blindness in consumer choice (Johansson, Hall, Sandberg, Dijksterhuis & Chater, in preparation). Here we have adapted an industry standard evaluation questionnaire used in focus group valuation of new products. Again, most of the participants fail to detect changes made to their evaluations (e.g. which product they find more natural, healthy, tasty, overall appeal, etc), and they also show the same type of preference reversal seen in other domains.
In the future of self-control (Hall, Johansson & de Léon, in press), we also present a general theoretical framework for how to address problems of self-control (for example, how to quit smoking) by using external cognition to increase our self-understanding. This line of argument is anchored in our empirical studies on choice blindness, as they show how we are often unaware of our fundamental lack of self-knowledge, while at the same time this lack can be alleviated by the use of external sources of information.
The studies I have conducted during my time as a Marie Curie fellow have shown that the choice blindness methodology can make a valuable contribution to the study of self-knowledge in consumer decision making, both in and outside the lab, and that the choice blindness situation (i.e. accepting and arguing for a choice not made) induces a preference change in a variety of experimental contexts, including consumer choice, monetary gambles and political opinion. These results are important in a number of different theoretical contexts (behavioural economics, decision making under risk, polling and voting behaviour, etc), as well as for applied consumer choice and marketing research.
Contact:
Petter.Johansson@lucs.lu.se
Website: http://www.lucs.lu.se(se abrirá en una nueva ventana)
