Predict Consumer Choice Research Articles

Predicting consumer choice from raw eye-movement data using the RETINA deep learning architecture

Predicting consumer choice from raw eye-movement data using the RETINA deep learning architecture

Leveraging genetic data for predicting consumer choices of alcoholic products

PurposeFood choices profoundly affect one's dietary, nutritional and health outcomes. Using alcoholic beverages as a case study, the authors assess the potential of genetic data in predicting consumers' food choices combined with conventional socio-demographic data.Design/methodology/approachA discrete choice experiment was conducted to elicit the underlying preferences of 484 participants from seven provinces in China. By linking three types of data (—data from the choice experiment, socio-demographic information and individual genotyping data) of the participants, the authors employed four machine learning-based classification (MLC) models to assess the performance of genetic information in predicting individuals' food choices.FindingsThe authors found that the XGBoost algorithm incorporating both genetic and socio-demographic data achieves the highest prediction accuracy (77.36%), significantly outperforming those using only socio-demographic data (permutation test p-value = 0.033). Polygenic scores of several behavioral traits (e.g. depression and height) and genetic variants associated with bitter taste perceptions (e.g. TAS2R5 rs2227264 and TAS2R38 rs713598) offer contributions comparable to that of standard socio-demographic factors (e.g. gender, age and income).Originality/valueThis study is among the first in the economic literature to empirically demonstrate genetic factors' important role in predicting consumer behavior. The findings contribute fresh insights to the realm of random utility theory and warrant further consumer behavior studies integrating genetic data to facilitate developments in precision nutrition and precision marketing.

가상현실(VR) 패션 스토어 환경에서의 소비자 휴리스틱 정보처리과정의 구조모델

Background/Objectives: This study aims to estimate the structural model of consumer heuristic information processing in the context of virtual reality(VR) fashion store. Specific objectives are to identify underlying dimensions of mental imagery, to examine the effect of mental imagery on telepresence, and to predict consumer choice heuristic for purchasing fashion products in the VR store environment
 Methods: The VR store of fashion brand X was selected as a stimulus in the VR experiment. A self-administered questionnaire was developed to measure mental imagery, telepresence and choice heuristics. By online survey, a total of 319 usable responses was obtained from consumers (female=169, males=150) who were aged from 20 to 39 years old.
 Results: The mental imagery was consisted of three factors, such as elaboration, haptic imagery, and spatial imagery. Estimated structural model showed that the elaboration and spatial imagery had positive effects on telepresence. Also, the telepresence influenced choice heuristics(e.g., intuition based heuristic and experience based heuristic), suggesting the mediating effect of telepresence between mental imagery and choice heuristics for fashion products at the VR store.
 Conclusion/Implications: This study provides insights into understanding mechanism of information processing in the VR experience. Also, managerial implications were discussed for applications of VR technology into strategic retail management in the fashion markets.

Classification of EEG Signals Using Neural Network for Predicting Consumer Choices.

EEG, or Electroencephalogram, is an instrument that examines the brain's functions while it is executing any activity. EEG signals to aid in the identification of brain processes and movements and are thus useful in the detection of neurobiological illnesses. Pulses have a very weak magnitude and are recorded from peak to peak, with pulse width ranging from 0.5 to 100 V, which is around 100 times below than ECG signals. As a result, many types of noise can easily influence them. Because EEG signals are so important in detecting brain illnesses, it is critical to preprocess them for accurate assessment and detection. The crown of your head The EEG is a weighted combination of the signals generated by the different small locations beneath the electrodes on the cortical plate. The rhythm of electrical impulses is useful for evaluating a broad range of brain diseases. Hypertension, Alzheimer, and brain damage are all possibilities. We can compare and distinguish the brainwaves for different emotions and illnesses linked with the brain by studying the EEG signal. Multiple research studies and methodologies for preprocessing, extraction of features, and evaluation of EEG data have recently been created. The use of EEG in human-computer communication could be a novel and demanding field that has acquired traction in recent years. We present predictive modeling for analyzing the customer's preference of likes and dislikes via EEG signal in our report. The impulses were obtained when clients used the Internet to seek for multiple items. The studies were carried out on a dataset that included a variety of consumer goods.

Sustainable success in the music industry: Empirical analysis of music preferences

Sustainability is not just a trend, but an important part of our everyday life including the satisfaction of human needs and preservation of a healthy business environment for present and future generations. The objective of this study is to provide an empirical approach for how to achieve the sustainable success in the music industry. As consumers’ utility toward a certain music product can be shifted up or down depending on the elements that constitute the music, we investigate the effect of musical elements on the consumer’s choice of music. We quantitatively measure the effects using hierarchical Bayesian logit choice model allowing for the individual heterogeneity. Based on the results, we find that utilizing musical components plays critical roles in understanding and predicting consumer choice. In addition, our findings suggest how music marketers can come up with a desirable configuration for music products. Sustainability in the music industry can be justified by whether musical components are well aligned, consistent with consumers’ preference.

Face to purchase: Predicting consumer choices with structured facial and behavioral traits embedding

Predicting consumers’ purchasing behaviors is critical for targeted advertisement and sales promotion in e-commerce. Human faces are an invaluable source of information for gaining insights into consumer personality and behavioral traits. However, consumer’s faces are largely unexplored in previous research, and the existing face-related studies focus on high-level features such as personality traits while neglecting the business significance of learning from facial data. We propose to predict consumers’ purchases based on their facial features and purchasing histories. We design a semi-supervised model based on a hierarchical embedding network to extract high-level features of consumers and to predict the top-N purchase destinations of a consumer. Our experimental results on a real-world dataset demonstrate the positive effect of incorporating facial information in predicting consumers’ purchasing behaviors.

Risks of Observable and Unobservable Biases in Artificial Intelligence Predicting Consumer Choice

Companies are increasingly adopting Artificial Intelligence (AI) today. Recently however debates started over the risk of human cognitive biases being replicated (and scaled) by AI. Research on biases in AI predicting consumer choice is incipient and focuses on observable biases. We provide a short synthesis of cognitive biases and their potential risk of being replicated in AI-based choice prediction. We also discuss for the first time the risk of unobservable biases, which affect choice indirectly, through other biases. We exemplify this by looking at looking at three prevalent, most frequently investigated biases in consumer behaviour: extremeness aversion, regret aversion and cognitive regulatory focus (prevention- versus promotion-focus). Based on a sample of 1747 respondents, through partial least squares structural equation modelling and significance testing, we show that regret aversion (unobservable bias) significantly reduces extremeness aversion (observable bias) and mediates the influence of cognitive regulatory focus (unobservable bias).

Predicting consumers’ choices in the age of the internet, AI, and almost perfect tracking: Some things change, the key challenges do not

AbstractRecent technology advances (e.g., tracking and “AI”) have led to claims and concerns regarding the ability of marketers to anticipate and predict consumer preferences with great accuracy. Here, we consider the capabilities of both traditional techniques (e.g., conjoint analysis) and more recent tools (e.g., advanced machine learning methods) for predicting consumer choices. Our main conclusion is that for most of the more interesting consumer decisions, those that are “new” and non‐habitual, prediction remains hard. In fact, in many cases, prediction has become harder due to the increasing influence of just‐in‐time information (user reviews, online recommendations, new options, etc.) at the point of decision that can neither be measured nor anticipated ex ante. Sophisticated methods and “big data” can in certain contexts improve predictions, but usually only slightly, and prediction remains very imprecise—so much so that it is often a waste of effort. We suggest marketers focus less on trying to predict consumer choices with great accuracy and more on how the information environment affects the choice of their products. We also discuss implications for consumers and policymakers.

Opportunities and Challenges for Using Automatic Human Affect Analysis in Consumer Research.

The ability to automatically assess emotional responses via contact-free video recording taps into a rapidly growing market aimed at predicting consumer choices. If consumer attention and engagement are measurable in a reliable and accessible manner, relevant marketing decisions could be informed by objective data. Although significant advances have been made in automatic affect recognition, several practical and theoretical issues remain largely unresolved. These concern the lack of cross-system validation, a historical emphasis of posed over spontaneous expressions, as well as more fundamental issues regarding the weak association between subjective experience and facial expressions. To address these limitations, the present paper argues that extant commercial and free facial expression classifiers should be rigorously validated in cross-system research. Furthermore, academics and practitioners must better leverage fine-grained emotional response dynamics, with stronger emphasis on understanding naturally occurring spontaneous expressions, and in naturalistic choice settings. We posit that applied consumer research might be better situated to examine facial behavior in socio-emotional contexts rather than decontextualized, laboratory studies, and highlight how AHAA can be successfully employed in this context. Also, facial activity should be considered less as a single outcome variable, and more as a starting point for further analyses. Implications of this approach and potential obstacles that need to be overcome are discussed within the context of consumer research.

Utilizing Case-based Reasoning for Consumer Choice Prediction based on the Similarity of Compared Alternative Sets

This study suggests an alternative to the conventional collaborative filtering method for predicting consumer choice, using case-based reasoning. The algorithm of case-based reasoning determines the similarity between the alternative sets that each subject chooses. Case-based reasoning uses the inverse of the normalized Euclidian distance as a similarity measurement. This normalized distance is calculated by the ratio of difference between each attribute level relative to the maximum range between the lowest and highest level. The alternative case-based reasoning based on similarity predicts a target subject's choice by applying the utility values of the subjects most similar to the target subject to calculate the utility of the profiles that the target subject chooses. This approach assumes that subjects who deliberate in a similar alternative set may have similar preferences for each attribute level in decision making. The result shows the similarity between comparable alternatives the consumers consider buying is a significant factor to predict the consumer choice. Also the interaction effect has a positive influence on the predictive accuracy. This implies the consumers who looked into the same alternatives can probably pick up the same product at the end. The suggested alternative requires fewer predictors than conjoint analysis for predicting customer choices.

Incorporating biometric data in models of consumer choice

ABSTRACTThe use of neuro-physiological data in models of consumer choice is gaining popularity. This article presents some of the benefits of using psycho-physiological data in analyzing consumer valuation and choice. Eye-tracking, facial expressions, and electroencephalography (EEG) data were used to construct three non-conventional choice models, namely, eye-tracking, emotion and brain model. The predictive performance of the non-conventional models was compared to a baseline model, which was based entirely on conventional data. While the emotion and brain models proved to be as good as conventional data in explaining and predicting consumer choice, the eye-tracking model generated superior predictions. Moreover, we document a significant increase in predictive power when biometric data from different sources were combined into a mixed model. Finally, we utilize a machine learning technique to sparse the data and enhance out-of-sample prediction, thus showcasing the compatibility of biometric data with well-established statistical and econometric methods.

Bayesian Revision vs. Information Distortion.

The rational status of the Bayesian calculus for revising likelihoods is compromised by the common but still unfamiliar phenomenon of information distortion. This bias is the distortion in the evaluation of a new datum toward favoring the currently preferred option in a decision or judgment. While the Bayesian calculus requires the independent combination of the prior probability and a new datum, information distortion invalidates such independence (because the prior influences the datum). Although widespread, information distortion has not generally been recognized. First, individuals are not aware when they themselves commit this bias. In addition, it is often hidden in more obvious suboptimal phenomena. Finally, the Bayesian calculus is usually explained only with undistortable data like colored balls drawn randomly. Partly because information distortion is unrecognized by the individuals exhibiting it, no way has been devised for eliminating it. Partial reduction is possible in some situations such as presenting all data simultaneously rather than sequentially with revision after each datum. The potential dangers of information distortion are illustrated for three professional revision tasks: forecasting, predicting consumer choices from internet data, and statistical inference from experimental results. The optimality of the Bayesian calculus competes with people's natural desire that their belief systems remain coherent in the face of new data. Information distortion provides this coherence by biasing those data toward greater agreement with the currently preferred position—but at the cost of Bayesian optimality.

Original brands in competition against high quality copycats

PurposeCopycat brands offering improved product quality pose serious challenges to original brands. This paper aims to provide a better understanding of why consumers prefer copycat brands with superior product attributes and how original brands can shift this preference back by strategically leveraging brand identity cues.Design/methodology/approachFour experimental studies test different types of brand identity cues that original brands can use to influence consumer preferences. Logistic and linear regression analyses analyze the effects.FindingsThe results systematically show the power of brand identity cues in helping original brands reduce share loss to copycat brands using superior product attributes. They also reveal the role of brand equity, conspicuous consumption and consumers’ tendency of using brands as status symbols in enhancing the effect of brand identity cues in the face of superior copycats.Research limitations/implicationsThis paper extends cue diagnosticity theory and the brand identity literature by showing the power of brand identity cues in predicting consumer choices of original brands.Practical implicationsThis paper provides useful guidelines for managers of original brands on how to effectively use brand identity cues to compete against copycats.Originality/valuePrior research focuses on how copycat brands’ characteristics influence consumers’ evaluations of copycats. These studies are limited, however, by their focus on cheap and low-quality copycats. The current paper examines the effects of brand identity cues and draws attention to the trade-offs consumers make when choosing between original brands and copycats offering superior product features.

Choice experiments are not conducted in a vacuum: The effects of external price information on choice behavior

Consumers carry a wealth of external knowledge and experience into a study that may affect choices in even the most well-crafted experiment. We derive a conceptual model showing how the level and variability of price perceptions “outside” the task at hand might affect choice behavior, and we conduct an experiment to test the model by manipulating information provided to respondents in a way to alter the “outside the experiment” perceptions. Mixed logit, error component models show that reference-dependent models best describe the choice data, and we find that consumer choice is significantly affected by the experiment manipulations. The predictions of the model are borne out in the data. As expected, price sensitivity is generally higher for price changes above a price reference point than below. Also as predicted, exogenously increasing external reference prices and exogenously reducing the variability in references prices both serve to increase the probability of making a purchase and reduce the probability of choosing “none.” These results have important implications for the design and practice of choice experiments, as they show that consumer choice is affected by perceptions of the level and variability of prices outside the experiment. More generally, the results highlight the need to understand the formation and distribution of reference prices when predicting consumer choice.

Neural antecedents of a random utility model

For over a decade, economists have sought to identify neural antecedents for economic theories. More recently, neuroeconomic work has sought to predict consumer choice using brain activations witnessed in non-choice, visual appraisals. This paper combines these two emerging strains of inquiry. Using functional magnetic resonance imaging (fMRI) data observed when consumers view a quality-differentiated food product labeled with different attributes, we seek to determine the predictive validity of the random utility model (RUM) often used in economic studies of consumer choice. Our fMRI data consist of changes in blood flow to the ventromedial prefrontal cortex (a brain region previously associated with value formation) observed when people saw low price and high labels and labels indicating high or low quality. We couple the fMRI data with data on 28 non-hypothetical choices made by each participant, which pitted higher priced, higher quality good vs. a lower priced, lower quality good. We find little evidence of a systematic difference in activation in brain areas thought to be associated with value formation when viewing high vs. low levels of attributes (prices and quality). However, differences in neural blood flow across participants related to quality (but not price) is significantly related to subsequent consumer choice both in- and out-of-sample, providing some qualified neuroeconomic support for the attribute-based RUM.

Consumption and climate change: Why we say one thing but do another in the face of our greatest threat

Abstract There is clear evidence that human beings have contributed to climate change through their patterns of consumption, and, it could be argued that, since we are part of the problem then we must be part of the solution. The apparently good news is that people report that they have very positive attitudes to environmentally-friendly products and they also consistently say that they are prepared to adapt their behavior to ameliorate the effects of climate change. However, numerous studies have found little behavior change on the part of consumers. This study investigates this critical issue experimentally. It does this by exploring whether self-reported attitudes to low carbon products, or alternatively implicit attitudes to such products (measured using an associative task and not requiring self-report), predict consumer choice of products varying on a range of dimensions including environmental consequences, in an experimental context where time for selection was also systematically varied. We found firstly, in line with previous research, that human beings have explicit and implicit attitudes that are not correlated. Secondly, in terms of brand choice, we found that consumers are particularly sensitive to both brand information and value in their selection of products, particularly under time pressure. Organic/eco brands are, however, much less favoured, especially under any time pressure, where processes that are more automatic prevail. Thirdly, color-coded carbon footprint information can influence choice even under time pressure but only for those consumers with a strong positive implicit attitude to carbon footprint. The conclusions from this research are that humans beings may well have a “divided self” when it comes to the environment and climate change, and this underlying “dissociation” in attitude (implicit versus explicit) might be critical to their behavior as consumers. This concept of the“divided self” may help us understand why relatively little has changed thus far with regard to more sustainable consumption, but might open up new lines of enquiry about how we might attempt to promote more sustainable consumption in the future.

Recent developments in identifying andquantifying emotions during food consumption.

Emotions and the consumption of food and beverages are inextricably intertwined. As the fields of sensory and consumer science seek to better conceptualize the consumer experience, interest in emotion measurement is growing. Emotions can provide key information to differentiate between products and predict consumer choice as well as give more detail about product perception. There are several emotion measurement instruments, including physiological methods and facial recognition, self-reported verbal emotion measurement and self-reported visual emotion measurement. This review discusses the purpose of measuring emotions, what is the definition of an emotion, what different instruments are available, and touches upon some promising research to deepen the connection between food and emotions. © 2016 Society of Chemical Industry.

A hierarchical choice modelling approach for incorporating customer preferences in vehicle package design

The use of customer preference models to evaluate the economic impact of design changes and new product introductions has become prevalent in the literature. However, existing approaches do not sufficiently address the needs of complex design artefacts, which typically consist of many subsystems and components designed and manufactured with significant autonomy. Characteristics of complex systems, such as heterogeneity of consumer preferences throughout the system hierarchy, multiple sources of information and qualitative consumer-desired attributes, have not been adequately addressed. In this work, we propose a hierarchical choice modelling approach for complex systems to model customer preferences for attributes throughout the system hierarchy, and to subsequently predict consumer choice behaviours. A system of hierarchical models is used to link the design attributes used for engineering design to the attributes used by consumers to choose among competing products. The model framework utilises Discrete Choice Analysis at the top level to model customer choices and Ordered Logit regression at the lower levels to model ordinal survey responses as a function of product attributes. An approach for combining choice data from multiple sources based on the Nested Logit methodology is developed. The framework is demonstrated on the vehicle occupant package case study.

Consumer Choice Prediction: Artificial Neural Networks versus Logistic Models

Conventional econometric models, such as discriminant analysis and logistic regression have been used to predict consumer choice. However, in recent years, there has been a growing interest in applying artificial neural networks (ANN) to analyse consumer behaviour and to model the consumer decision-making process. The purpose of this paper is to empirically compare the predictive power of the probability neural network (PNN), a special class of neural networks and a MLFN with a logistic model on consumers' choices between electronic banking and non-electronic banking. Data for this analysis was obtained through a mail survey sent to 1,960 New Zealand households. The questionnaire gathered information on the factors consumers' use to decide between electronic banking versus non-electronic banking. The factors include service quality dimensions, perceived risk factors, user input factors, price factors, service product characteristics and individual factors. In addition, demographic variables including age, gender, marital status, ethnic background, educational qualification, employment, income and area of residence are considered in the analysis. Empirical results showed that both ANN models (MLFN and PNN) exhibit a higher overall percentage correct on consumer choice predictions than the logistic model. Furthermore, the PNN demonstrates to be the best predictive model since it has the highest overall percentage correct and a very low percentage error on both Type I and Type II errors.

Understanding caregivers' help-seeking for Latino children's mental health care use.

Latino children have persistent low rates of mental health service use. Understanding the factors that influence caregivers' decisions about whether to use mental health care for their children can help explain why. The objective of this study was to investigate the factors reported by the primary caregiver that could help classify Puerto Rican children into users versus nonusers of mental health services and mental health versus school sector care, among users. Data were collected from a random Puerto Rican community sample of caregiver-child dyads. Version-IV of the Computerized Diagnostic Interview for Children (DISC) was used to assess psychiatric disorders in children. The Service Assessment for Children and Adolescents (SACA) was used to examine the types of services used for mental health problems. The Classification and Regression Tree (CART) approach was used to develop a simple model simulating caregivers' decision-making around taking children for mental health care and the setting for care. The classification model of use versus no use of mental health service suggested 3 significant predictors: child's level of impairment, parental concern, and child's difficulty in performing schoolwork. The classification model of sector of care, mental health versus school setting, identified 1 significant predictor, any disruptive disorder diagnosis. : Assisting caregivers in linking a child's impairment with need for mental health care might be a mechanism to reduce children's unmet need. Approaches such as CART, used to identify factors predicting consumer choices in marketing, might be useful to select strategies for social campaigns targeted toward decreasing unmet need.