DOI: 10.31038/PSYJ.2020215

Abstract

Respondents in two studies evaluated systematically varied combinations of appearance descriptors about another person to be matched with them in a dating site (Study 1), or behavior-description about another person with whom they are currently in a relationship (Study 2.)Study 1 on physical appearance showed features which drove positive responses. Study 2 on relationships did not show features which drove positive responses. Neither study showed dramatically opposite mind-sets of the type usually seen when people evaluate products, services, or situations.  These results suggest that respondents have an easier job judging external stimuli, rather than stimuli appropriate to their own personality, supporting the proverbs which talking about the blindness people have towards their own behavior.

Introduction

Love, whether romantic or married, licit or illicit, experienced by the young or by the old, fascinates and has fascinated for millennia. One need only poetry to get a sense of the fascination with love.  The world is changing, however, and dramatically so [1] suggest that “the solidity and security once provided by life-long partnerships has been ‘liquefied’ by rampant individualisation and technological change… internet dating is symptomatic of social and technological change that transforms modern courtship into a type of commodified game [2] bring the topic even more up to date by investigating the increasingly popular mobile dating apps, technologies that up to then had received little academic investigation, despite their growing popularity.

The two studies presented here are not meant to be definitive, or even deeply exploratory. Rather, they present a way to understand as aspect of human feelings, love and perhaps marriage, from the point of view of the inside of the mind.  A word of explanation is appropriate here. A great deal of the scientific literature on love and its ramifications is given over to the statistics of love, to the frequencies of behaviors, and to the depth of feelings as measured by scale. From these metrics, one gets a sense of the nature of this universal phenomenon of love. The ordinary person, and indeed the therapist dealing with issues of love, does not understand love from the point of view of numbers and statistics, but rather from the world view of experience, the internal experience, what is going on in the mind of the people who are being studied or helped.  It is the idiographic, the personal, which makes the world of love so interesting. Each person has a story, and when generally well-told, the story is interesting.

The philosophical and methodological contribution of Mind Genomics

A glance at the scientific literature in virtually any field today shows the prevalence of the hypothetico-deductive method (henceforth abbreviated as HDM). The world view of the HDM is that knowledge in science proceeds by creating hypotheses, whichcan then be confirmed or falsified.  Confirmation of a hypothesis does not mean that the hypothesis accurately and correctly describes how nature is working. In contrast, disconfirmation, falsifying, suffices to disprove the linkage between hypothesized cause and effect.

The HDM leads to a certain approach in the literature, in which new studies are grounded in the results of previous studies. That is, the previous studies, having been published, are assumed to present scientific evidence which ‘call for confirmation or falsification.’  As a result, science can be seen in the literature to process in a stepwise, cautious, quite conservative manner, with the previous studies leading to the current studies. In common research parlance, many studies are done to ‘plug holes in the literature.’ That statement is not disparaging, but simply the language that is used.

In contrast to the approach of conventional science, which can be called ‘nomothetic,’ to recognize the desire for nomos, law, is the idiographic, the study of the individual, and in many ways the study of the richness of the experience of an individual.  Whereas the nomotheticmight be filled with statistics and analyses, the idiographic is often more interesting reading, dealing with the specificexperience.  There is no issue involving nomos, laws, other than observations of repeating pattern. Rather, there is the focus on the individual, on the richness of experience.

Design set up and analysis

The pair of studies reported in this paper comprise the authors’ attempt to combine the idiographic and the nomothetic, using a method that is best conceived of as an experiment. The respondent is presented with systematically varied combinations of messages, assigns a rating, and from that simple set of activities the underlying criteria used by the respondent to make the decision are uncovered. We apply the method to relationships, at the level of the physical attractor, and at the level of behavior in a marriage.

The Mind Genomics studies are set up in a straightforward manner which allows the researcher to understand the ‘mind of the respondent’ in terms of what messages drive positive versus negative judgments. The origin of Mind Genomics is traceable to the Socratic tradition of question and answer, from which truth emerges. The origin is also traceable to experimental psychology, and the perception of patterns in noise. Previous papers in a variety of journals present the approach, which is herewith summarized by the series of steps [3, 4].  The approach, based a combination of statistics [5], consumer research [6], and intuitive, ‘System 1 Thinking’ [7], provides an efficient way to understand the mind, rapidly (hours), inexpensively, with scientific rigor, and with actionable results.

1. Select the topic, ask four questions which ‘tell a story’ and provide four ‘answers’ to each question.  It is the answers, short phrases, that are mixed to become the test stimuli. The questions are used to elicit the answers. The questions themselves never appear.
2. Combine the answers into vignettes, combinations, without any connectives. The vignette comprises two, three, or four answers, at most one answer from a question. The vignettes are created according to an experimental design, a set of recipes, in this case 24 vignettes or pre-set combinations. Each answer appears five times, and absent 19 times in the set of 24 vignettes. Each respondent evaluates a totally unique set of 24 vignettes, different from the vignettes evaluated by any other respondent [8] Many of the vignettes will be incomplete by design, a fact which does not bother the respondent, although occasionally is the source of discomfort to the researcher or the researcher’s ‘client.’ The ingoing belief causing such discomfort is that the respondent ‘cannot possibly’ rate the incomplete vignette, a discomfort proved wrong again and again by the successful experiments with such incomplete vignettes.
3. The rating question is 5, 7, or 9-point category or Likert scale. The scale is anchored at both ends, and in some cases may be anchored at every point.  Prior to the analysis the scale is bifurcated to create two scales, a negative and a positive. When the focus is on the positive aspect (e.g., interested, or successful marriage) the scale will be truncated so that the top 2 scale points on the 5=point scale will be assigned the value 100, and the remaining 3 lower scale points will be assigned the value 0. In contrast, when the focus is on the negative aspect (not interested, marriage will be a failure), the lowest two scale points will be assigned the value 100, and the higher three scale points will be assigned the value 0.
4. The method of OLS, ordinary least-squares regression, relates the presence/absence of the 16 elements to the ratings. The experimental design allows the OLS regression to be used to create models for each respondent, in preparation for clustering and segmentation.
5. The pattern of 16 coefficients for the individual respondents in a study is used to create two and three cluster groups. A cluster or mind-set is defined as a group of respondents who show a similar pattern of their coefficients for the 16 elements, suggesting a similar way of thinking about the topic. There is no reason to assume that there is any a priori relation between WHO a respondent is as determined by the classification questionnaireand how the respondent THINKS as determined by the clustering.
6. Prior to the actual experiment, the respondent profile himself or herself using a short questionnaire, to determine gender, age, and then their membership in a third group, that group being the one of interest to the researcher. For these early stage studies, the focus of subgroups will be gender differences, and then mind-sets.

Study 1 – Dating site – before the relationship

Study 1 was occasioned by the inquiry of two female teenagers in Montenegro about how one would use Mind Genomics to understand what people are looking for in relationships.  Beyond the actual experiment in Mind Genomics is the nature of the test stimuli, and what that suggests about how the researcher ‘thinks.’ Science typically focuses on the answers to the questions, attempting to understand the field by the way answers fit together to create a larger picture. Mind Genomics moves one step forward with cognitively meaningful test elements.  Not only are the patterns of responses to the test elements relevant, but the pattern of what test elements (answers to questions) becomes important.  For this study The elements are all simple physical descriptions of the person.  Table 1 shows the 16 answers, sorted in terms of the degree of interest each generates when part of a vignette.

Table 1. Dating-site models relating the presence/absence of elements to the binary transformed rating of interested

The respondents were US residents, members of an on-line panel (Luc.id), and accustomed to participating in on-line surveys. The respondents are compensated for their participation. The age range specified was 18–35. The panel comprised 16 males and 14 females. No additional information beyond gender, age, and relationship status was collected, ensuring that there was no information which could be used to identify a respondent.  Table 1 shows the summary results for Total panel, gender, and then two emergent mind-sets

1. Additive constant = estimated percent of responses 4–5 in the absence of elements. The additive constant is a purely theoretic, estimated parameter. It is a good baseline, however. The additive constant for the total panel, 51, suggests that in the absence of elements, about half of the responses should be ‘interested’.  The males show a slightly higher additive constant, 55, whereas the females show a slightly lower constant, 43.    The two mind-sets also show different additive constant (48 for MS1 vs 60 for MS2)
2. The shaded cells show coefficients of 8 or higher, corresponding to strong performers, with a statistically significant coefficient (P, 0.05). The total panel shows no strong performing elements, but the gender subgroups and MS1 do show strong performing elements

   Males (n=16, 12 in MS1, 4 in MS2) – respond to nice teeth, soft voice, and a wide smile

   Females (n=14, 8 in MS1, 6 in MS2) – respond to green eyes

   Mind-Set1 (MS1, n=20, 12 males 8 females) – wants green eyes, big eyes, soft voice, nice teeth and a kind smile

   Mind-Set 2 (MS2, n=10, 4 males, 6 females) – shows a higher additive constant, but nothing specific.

When we change our focus from what interests the respondent to what turns off the respondent, we find a totally different pattern.  Recall that the scale was bidirectional. By coding ratings of 1–2 as 100, it becomes straightforward to discover hat is a ‘turn-off’ to the respondent.

Table 2 shows a much richer pattern of turn-offs, compared to Table 1 which showed the pattern for ‘turn ons.’ The additive constant is lower, around 30, meaning that in the absence of elements, we expect 30% of the ratings to be negative (1 or 2 on the five-point scale.)

Table 2. Models relating the presence/absence of elements to the binary transformed rating of disinterested

Stronger patterns emerge from the ‘turnoffs’

The total panel feel that the extremes are turn-offs, especially voice. Some of these may be gender related.

Males are more critical than are females, and Mind-Set 1 is more critical than Mind-Set 2.

The Mind-Sets show different patterns of turn-offs, with a few common turn-offs (petite height, high pitched voice, tall as a basketball player

Petite height, thigh pitched voice, and tall as a basketball player appear to be universal turn-offs.

The final analysis for the dating study looks at the Consideration Time, defined at the number of seconds elapsing between the appearance of the vignetteon the screen and the response.  Consideration times lasting longer than 9 seconds were brought to 9 seconds, assuming the respondent was multi-tasking.  The modeling created an equation without an additive constant, based upon the premise that in the absence of elements there is no response.

We assume that the consideration time reflects the amount of time need to read the element and process it as part of the decision. The consideration time is not good or bad, but simply reflects the amount of processing going on.

A number of the longer consideration times tend to be those associated with negative interest, as if it were taking the respondent extra time. Examples are ‘Tall as a basketball player’ and ‘petite height.’

On average, males and females show the same average Consideration Time. In contrast, Mind-Set 1 (MS1) shows a long Consideration Time (average =1.0 second), where as Mind-Set 2 (MS2) shows a  short Consideration Time (average = 0.6 seconds)

Study 2 – Marriage in trouble

The second study deals with marriage in trouble, and the likelihood that it will last.   Marriage is clearly changing. As [1] suggested, “the solidity and security once provided by life-long partnerships has been ‘liquefied’ by rampant individualisation and technological change… internet dating is symptomatic of social and technological change that transforms modern courtship into a type of commodified game.”  The phenomenon known as the gray divorce (after age 50) has exploded in frequency, as adults later in life decide to end their marriage, and live as singles [9, 10] One out of every four divorces in the United States is a gray divorce, presumably because the partners want to experience personal growth. Nonetheless, the commitment model is still quite strong in most marriages, a commitment to help each other, based upon binding, romantic love. It is only when the personal strains between the members of the couple become so great as to be unbearable that the gray divorce occurs [11].

Divorce is not the only problem. So is the marriage itself, which world-wide appears to be occurring at a later age. For example, in Hong Kong, a set of interviews by [12] suggest that rather than divorce the marriage is postponed, to a great degree because of the difficulties associated with transitions from school to work.

To fully plumb the topic of marriage, divorce, relationships would require a far large undertaking than one or two experiments. Nonetheless, the Mind Genomics approach might well shed some light on some of the mind-sets involved, and the different criteria used to judge the potential failure versus success of a marriage. As the data will show, the cause of success versus failure in a relationship elude simple patterns, a ‘first’ for Mind Genomics, which generally finds clear patterns and a small number of meaningful, interpretable mind-sets.

The second study in this pair concerned the evaluation by respondents as to the future of a marriage by four variables, the four questions (reasons for marriage, spouse intimacy style, spouse communication style, spouse financial style.) Discussions with those in the counseling field suggested these four categories, which were then fitted into the Mind Genomics format.

The same set-up approach was used, comprising the topic, the four questions, and the four answers to each question. The Mind Genomics system, created in the form of an input template, enables the research to create the study quickly, once the questions and answers have been created.  The   rating question was: What is the future of this relationship in the next 6 months:  1 = Split-up … 5 = Improving

Table 3. Models relating the presence/absence of elements to the Consideration Time (seconds used to process the information and make a judgment)

The set-up of this first study on relationships, the dating site, revealed the way teens think about the other sex.  The test stimuli created by the researchers suggest that they perceive the other gender in terms of physical properties. The longer Consideration Times occur for those descriptions of physical attractiveness which are ambiguous, and not the typical response of what ‘attractive’ is.

The respondents comprised 34 panelists, once again supplied by Luc.id, all 35 and older.

Males (n=13, 6 in MS1, 7 in, MS2)

Females (n=21, 11 in MS1, 10 in MS2) – More optimistic than males

Mind-Set 1 (MS1 n = 17, 6 males, 11 females)- More optimistic than MS2. For Mind-Set 1, the more optimistic, some answers, especially spouse communication, fail the ‘meaning test.’  This departure is worthy of deeper investigation because both positive and negative ‘spouse communication styles’ are performing strongly, and positively.

Mind-Set 2 (MS2 n = 17, 7 males, 10 females)

The data allowed for the same analysis as Study 1, specifically what drives positive feelings (marriage improving, ratings of 4–5 transformed to 100, ratings 1–3 transformed to 0), what drives negative feeling (split up, ratings 1–2 transformed to 100, ratings 3–5 transformed to 0), and the Consideration Time.

Despite the various strong performing elements, and thus a variety of elements which drive improvement, there is no clear pattern for any group. We can conclude that there is no clear understanding of what will improve the marriage, knowing the situation. (Table 4- Table 6)

Table 4. Models relating the presence/absence of elements to the binary transformed rating of marriage improves in six months

When we look at the negative side, we see that the belief in a split is about 37% to 40% as a baseline, just a little lower than the baseline for success.   Furthermore, there are only threestrong elements which drive a breakup, two dealing with finances, one dealing with bonding style.

Table 5. Models relating the presence/absence of elements to the binary transformed rating of split up in six months

Our final analysis deal with Consideration Time, this time for the future of a marriage (Table 6).  The consideration times are fairly long, especially intimacy styles.  The shortest Consideration Time is spousal financial habits.  The Consideration Time suggests a basic fascination with intimacy, and a basic lack of fascination with financial habits.  Fascination does not mean positive or negative, but simply ‘engagement time, i.e., time spent considering the element when making a decision.

Table 6. Models relating the presence/absence of elements to the Consideration Time

These data suggest that the complexities of marriage are quite different from the complexities of a relationship, and that it may be simply impossible to predict the likelihood of a breakup. The patterns are not clear at all in the way they are clear using Mind Genomics for so many other topic areas

Overall discussion

The topic of love, from early attraction to its death in the end of a marital (other) relationship is a topic fascinating everyone, from writers and poets to sociologists, psychologists, marketers, and so forth. The millions of articles each year on attraction, relationship, love, and the heartbreaks which ensure are silent witness to the preoccupation of people with love.

The preliminary data in this pair of experiments suggests that it is easier to deal with the physical aspects which drive attraction. When the topic turns to relationships, especially marital relationship and their future under stressful conditions, Experiment 2 suggests a totally different story, one in which there is a difficulty inherent in uncovering meaningful, interpretable patterns. Whereas most Mind Genomics studies reveal easy-to-label mind-sets, Experiment 2 in our work reported here suggests that the mind-sets are not clear. There are probably many more mind-sets, so perhaps we are lacking the requisite base size. Yet, for the same base size of 30–35 respondents, other studies reveal quite different, and easy to label mind-sets.

Acknowledgment

Attila Gere thanks the support of the Premium Postdoctoral Researcher Program of the Hungarian Academy of Sciences.

These data suggest that the complexities of marriage are quite different from the complexities of a relationship, and that it may be simply impossible to predict the likelihood of a breakup. The patterns are not clear at all in the way they are clear using Mind Genomics for so many other topic areas

References

1. Hobbs Mitchell, Stephen Owen, Livia Gerber (2017) Liquid love? Dating apps, sex, relationships and the digital transformation of intimacy.” Journal of Sociology 53: 271–284.
2. Albury K, Burgess J, Light B, Race K, Wilken R (2017) Data cultures of mobile dating and hook-up apps: Emerging issues for critical social science research. Big Data & Society 4.
3. Moskowitz HR, Gofman A, Beckley J, Ashman H (2006) Founding a new science: Mind genomics. Journal of sensory studies 21: 266–307.
4. Moskowitz HR, Gofman A (2007) Selling blue elephants: How to make great products that people want before they even know they want them. Pearson Education.
5. Luce RD, Tukey JW (1940) Simultaneous conjoint measurement: A new type of fundamental measurement. Journal of mathematical psychology 1: 1–27.
6. Green PE, Srinivasan V (1990) Conjoint analysis in marketing: new developments with implications for research and practice. The journal of marketing 54: 3–19.
7. Kahneman D (2011) Thinking, fast and slow. New York: Farrar, Straus and Giroux.
8. Gofman A, Moskowitz H (2010) Isomorphic permuted experimental designs and their application in conjoint analysis. Journal of Sensory Studies 25: 127–145.
9. Brown SL, Wright MR (2017) Marriage, cohabitation, and divorce in later life. Innovation in Aging 1.
10. Grover S, Helliwell JF (2019) How’s life at home? New evidence on marriage and the set point for happiness. Journal of Happiness Studies 20: 373–390.
11. Crowley JE (2019) Gray divorce: Explaining midlife marital splits. Journal of women & aging 31: 49–72. [Crossref]
12. Ng T (2019) Is marriage in decline: A qualitative study of perceptions of young people on marriage in Hong Kong, China. Asian People Journal 2: 126–137.