Overview of the Neopic™ and Neo-Archetype™ engines

23rd August 2019

Author: Nigel Marlow

 

Advances in neuroscience during the past decade (e.g. Bargh et al, 2002[1]; McLure et al, 2004[2], Zaltman, 2003)[3]have demonstrated beyond any doubt that the majority of human behaviour is driven by brain processes that operate below conscious awareness.

The processes influencing this involuntary part of thinking, feeling and behaviour, include memory, attention, perception, attribution of meaning, emotional responses and decision making.

All these subconscious influences shape our behaviour and regularly impact on the choices we make.

As professional researchers in the marketing and branding arena, we need to accept the challenge of understanding these subconscious processes more fully in order to predict consumer behaviour accurately. The singular reliance on traditional research approaches (e.g. focus groups, surveys etc.) only accesses conscious material. This equates to collecting data from the ‘tip-of-the-iceberg’ when compared to the rich seam of emotional and motivational material that is available in the subconscious.

Any professional researcher (e.g. marketer, business psychologist etc.) who has been tasked with eliciting data from people, will be familiar with the potential ‘cloudiness’ associated with the typical responses (whether collected qualitatively via interviews and/or focus groups or quantitatively from Likert-style surveys).  The uncertainty hidden behind the research results potentially arises from any or all of the following:

  • participants respond even when they don’t know what they think
  • people may not tell the truth (e.g. for fear of what others might think- social desirability)
  • people are bad at predicting their own behaviour
  • consumers rarely know the true reason whythey have behaved in a certain way
  • individuals often ‘second-guess’ what the researcher wants to hear

In addition, an increasing number of researchers[4], have questioned the reliability and therefore the validity of traditional measure that use Likert-scales. The Likert ‘scale’is not a true scale in the mathematical or statistical sense and should therefore not be analysed by inferential statistics applied to its aggregated data. The Likert suffers also from an inherent central tendencybias, which questions the meaningfulness of any distributional markers such as means and standard deviations.

At Innovationbubble, our response to these widely known problems has been to invest in searching for new ways in which to elicit more reliable and therefore more valid information for our clients.

The human brain processes images 60,000 times faster than it does text[5].  Marketers have recognised this huge differential in the speed of processing between text and images, and with the explosion of the use of Smartphones[6]visual marketing (images and video) is now the dominant approach in the digital marketing business.

In addition to this advantage of speed of processing, images also have the potential of being a much more emotional and compelling medium. “Marketing without words enables you to tap into the underlying power of images – emotions. Images cause people to react, and those reactions are emotional ones.”[7]

This practical evidence for the qualitative differences between the processing of images and text from the world of Marketing is supported by academic psychological research. For example, images are recalled more easily than word stimuli[8], are easier to associate with other words or images[9]and are decoded for meaning more quickly than words[10].Other research has shown thatpeople can remember more than 2500 images with at least 90 percent accuracy for days after initial exposure[11].

To summarize, it is evident that visual content reaches an individual’s brain in a faster and more understandable way than textual information. The human brain is hardwired to recognize and make sense of visual information more efficiently than text-based data and images evoke emotional responses more powerfully than words.[12]

Given this image-superiority effect, the R&D team at IB undertook the task of developing a set of new measures, which were based upon the revolutionary approach of replacing the text-descriptors and Likert-scales traditionally found in current survey instruments, with pictures and single-word stimuli.

The resultant measures, Neopic ™ and Neo-Archetype™, have no Likert scales, which have been replaced with a unique image- based calibration system. The images used were selected using a construct-rational strategy[13]. NEOPIC™ scores have positive correlations with the traditional NEO5 questionnaire[14].  The results correspond to those found by Paunonen et al (2001)[15]when developing their nonverbal clinical version of the NEO5.

Other researchers have since developed different versions of image-based questionnaires. These include Visual DNA (2008)[16], having produced a variant of the NEO5 and Leutner et al (2017) who have developed a new creativity measure.[17]

However, both the Paunonen and Leutner measures have retained text-based questions, and the images used only appear as a replacement for the numbers in the Likert scales. For example, an image-based scale of ‘involvement’ might show pictures depicting “toe-in-the-water”, “paddling” and “diving-in” etc. as the graduated possible responses replacing “1” , “2” and “5” in the original scale.

By contrast, the NEOPIC™ and Neo-Archetype™ engines are entirely image based. Their designs were developed specifically to elicit spontaneous responses, which are more likely to represent the true underlying thoughts and feelings of individuals. This reduces the biases inherent in Likert scale responses, which are contaminated with the ‘masking’ of social desirability etc.

The NEOPIC™ and Neo-Archetype™ images are also more likely to elicit emotional responses, and therefore reflect the complexity of an individual’s personality more fully than the rather ‘Spartan’ Likert-Item response format.

The NEOPIC ™ and Neo-Archetype™ processes are more engaging and motivating. (It is well-known that traditional mathematical scales quickly become boring and tiring after the first 100 items)! However, the NEOPIC ™ and Neo-Archetype™ only take 2-3 minutes to complete and therefore the data collected is likely to be more reliable without the influences of tedium and fatigue.

The use of images in the NEOPIC™ and Neo-Archetype™ processes also has the advantage of producing a holistic measure of personality, which is more useful in organisational and marketing research. The output is a ‘fuzzy-sketch’ or prototype of the respondent[18], which mirrors how we perceive and evaluate each other.  As actors in a social scene, we normally perceive ‘personality’ as a latent emergent property; ‘personality’ is a ‘whole’ and not a disjointed bundle of separate traits or dimensions.

Neopic ™ and Neo-Archetype™ do not rely on mathematical data and reject the psychometric approach that assumes numbers are able to accurately measure psychological concepts such as ‘personality’. Such emergent properties of an individual (including ‘intelligence’) are at best ‘fuzzy’ and differences in numbers scored may not represent a real or meaningful difference between individuals.

The Neo-approach disengages from the assumption of a 1:1 correspondence between the score on a mathematical scale and the respondent’s personality characteristics. In its place, the Neo-process allows an individual’s preferences and motivations to be expressed more naturally and authentically. The respondent offers personal information more ‘freely’ in the spirit of self-discovery. The feedback focuses on reflecting the person’s characteristics for the benefit of the respondent, rather than providing a ‘measure’ for someone else’s evaluation.

The Neopic™ and Neo-Archetype™ are less stressful or arduous than traditional questionnaire formats. Participants can actually enjoy taking the test and are more fully engaged with the quiz-style format. It is much quicker to complete than formal tests and its spontaneity is more likely to elicit ‘honest’ responses that originate directly from the subconscious without having been ‘thought-about’.

 

APPENDIX

The testing of parallel forms of the NEO5

We have compared the traditional format of a personality test (NEO5) with the Neopic™ design. The following paragraphs briefly describe the theoretical foundations and validation process used in the development of Neopic™ and Neo-Archetype ™.

BACKGROUND

There are many personality theories and approaches which each have a range of methods for evaluating individual characteristics. The dominant approach has become ‘Trait Theory’, which postulates that personality consists of several separate personality traits or dimensions; popular examples of these being introversionand extraversion.

The dimensions or traits are usually measured using questionnaires or inventories based upon Likert-scales. The outcome is a set of mathematical scores for individual traits that can be compared with the scores of other individuals and/or the averages of populations. These types of questionnaire are known collectively as psychometrictests and have certain statistical properties that allow their mathematical reliability and conceptual validity to be evaluated.[19]

The statistical robustness of these questionnaires has meant their wholesale adoption by HR and Marketing Departments for the use in the selection, segmentation and other research purposes.

Currently, the most popular personality measure in use is the NEO5, which is based upon a model of personality that has five dimensions (Costa and McCrae, 1992); these are Neuroticism, Extraversion, Agreeableness, Conscientiousness and Openness.

*We are set-up to allow you to experience the NEO5 and our Neopic™ versions online.

A shortened60-item version of the psychometric version of the NEO5 is available on request for you to try. After completing the test, we would like you to reflect on your experience and the outcome (feedback).

  • How accurately do you think it describes your personality?
  • How much consideration did you give to items before responding?
  • How interesting did you find completing the questionnaire?
  • Do you think the 120-item version of the questionnaire will be more accurate?
  • How long did it take you to complete the questionnaire?
  • Were you honest with all your answers?
  • Overall, what was you’re feeling about the questionnaire?

 

Now we would now like you to try our isomorphic version of the NEO5 test that you have just completed. We can arrange for you to try the NEOPIC™ version on request.

We are confident that your experience of taking the Neopic™ version will be considerably more positive than that of completing the traditional format of the NEO5. Furthermore, we expect that you will find the feedback from the new version more accurate and engaging.

We think that the NEOPIC ™ version will:

  • More accurately express your personality – how you feelabout yourself
  • Be more quickly completed
  • Be more engaging
  • Be more interesting
  • Be more thought provoking

 

Brief outline of the Neopic™ development

Images were chosen to replace the NEO5 items based upon our knowledge of projective techniques. The images were also selected to elicit a balanced range of both positive and negative feelings. Participants responses are designed to indicate whether they are attracted towards a particular image or not.

The range of valences for the images was arranged so that selected images contributed positively or negatively to a given personality dimension. An internal algorithm then matched the image-selection to a corresponding behavioural profile.

Several pilot studies were run to adjust the menu of images and the connecting algorithm that selects the appropriate feedback.

Our initial validation studies involved individuals completing the standard NEO5 (short-form). Respondents were then asked to take the NEOPIC™ version.

For testing purposes, the dimensions underlying the NEOPIC™ were allocated numerical value.

The studies involved approximately 350 participants over several weeks. Data was statistically analysed and respondents ‘comments recorded.

The results showed a moderate positive correlation, (see Chamorro-Premuzic and Ahmetoglu, 2013) between the NEO5 and NEOPIC ™ versions which fell between 0.30 and 0.40, the significance level was p<0.05.

This positive correlation demonstrates that the NEOPIC™ is tapping into the NEO5 personality concept and it is validated by its statistical significance; there is only a 5% chance that the result has occurred by luck and we can be 95% confident of the positive relationship between the two versions of the personality measurement.

 

Further technical advantages of the NEOPIC ™ process:

  • The responses to the NEOPIC™ images are mainly based on a projective process and therefore express more of the underlyingtruepersonality rather than the ‘mask’ of what is socially desirable.
  • Images are also more powerful in eliciting emotionalresponses in addition to those arrived at by ‘cold’ consideration. The resulting ‘measure’ is therefore more likely to reflect the complexity of an individual’s personality.
  • The NEOPIC™ process requires less cognitive effort and taps into System 1 responses that are more automatic and immediate than those arrived at with more consideration. These responses are therefore more likely to be a truer representation of the thoughts and feelings of the individual.
  • The interactive nature of the NEOPIC™ process reduces the likelihood of irrelevant data being collected because respondents are becoming tired or bored or both! Individuals completing the NEOPIC™ are more likely to remain engaged with the material and therefore more motivated to portray a truer picture of themselves.
  • The boundary of meaning of the images is more indistinct than their Likert-scale counterparts. Therefore it is more difficult to measure each specified personality trait in a clear and distinct way. However, this is how our ‘personality’ is perceived. We do not think of others as a bundle of distinct dimensions, rather we hold an overall impression of them. It is a sort of ‘fuzzy’ image or prototype of that person[20], which changes shape and colour and focus according to the environmental contexts.
  • These descriptions of ‘fuzzy-personality’ are another way of describing the concept of ‘archetypes’ – the mental models of our ‘theatre-personas’ and (hidden) aspirations.

The NEOPIC™ and Neo-Archetype™ are measuring a more holistic concept of ‘personality’ or persona and are providing an ‘image’ of the total person rather than a loose collection of individual dimensions, expressed in soulless numbers.

 

 

REFERENCES

Chamorro-Premuzic, T. & Ahmetoglu, G. (2013) Personality 101, NY: Springer

Costa, P. T., Jr., & McCrae, R. R. (1992), “Four ways five factors are not basic”: Reply. Personality and Individual Differences, 13, 861- 865

[1]Bargh, J. A., Gollwitzer, P. M., Lee-Chai, A., Barndollar, K. & Troetschel, R. (2001); The automated will: Non-Conscious activation and pursuit of behavioural goals; Journal of  Personality and Social Psychology, Vol.81(6), December 2001, 1014-1027.

[2]McClure S. M., Li J., Tomlin D., Cypert K. S., Montague L. M. & Montague P. R. (2004).”Neural Correlates of Behavioural Preference for Culturally Familiar Drinks”; Neuron, Vol. 44, No. 2, October 2004, 379-387.

[3]Zaltman, G. (2003). How customers think: essential insights into the mind of the market. Boston, MA :Harvard Business School Press

[4] For examples see: Carifio J, Perla R. Resolving the 50-year Debate Around using and Misusing Likert Scales, Med Educ. 42(12): 1150-1152, 2008; Dawes J. Do Data Characteristics Change According to the Number of Scale Points Used? An experiment Using 5 point, 7 point and 10 Point Scales, Int. J Market Res 51(1), 2008; Knapp TR. Treating ordinal scales as interval scales: an attempt to resolve the controversy. Nursing Res 39(2): 121-123, 1990; Allen IE, Seaman CA. Likert Scales and Data Analyses. Quality Progress 40(7):64-65, 2007.

[5]See Ritu Plant (2015); http://www.business2community.com/digital-marketing/visual-marketing-pictures-worth-60000-words-01126256#55XRVCzFQbwekOQF.97

[6]Over 80% of adults worldwide own a Smartphone

[7]See Apu Gupta (2017); https://www.forbes.com/sites/onmarketing/2013/07/02/the-shift-from-words-to-pictures-and-implications-for-digital-marketers/#478410c8405a

[8]Defetyer, M. A.; Russo, R.; McPartlin, P. L. (2009), “The picture superiority effect in recognition memory: a developmental study using the response signal procedure”; Cognitive Development. 24: 265–273.

[9]Hockley, W. E. (2008), “The picture superiority effect in associative recognition”, Memory & Cognition, 36(7): 1351–1359.

[10]Hockley, W.E.; Bancroft, T. (2011), “Extensions of the picture superiority effect in associative recognition”. Canadian Journal of Experimental Psychology, 65(4): 236–244.

[11]See Peter temple (2017); http://presentationsforbusiness.com/powerpoint/why-visuals-are-more-powerful-than-words/

[12]See eyeQ (2017); https://www.eyeqinsights.com/power-visual-content-images-vs-text/

[13]Jackson, D. N. (1970), A sequential system for personality scale development, Current Topics in Clinical and Community Psychology, Volume 2, New York; Academic Press

[14]Chamorro-Premuzic, T. & Ahmetoglu, G, (2013), Personality 101, NY; Springer

[15]Paunonen S V, Ashton M C, and Jackson, D N, (2001), Nonverbal Assessment of the Big Five Personality Factors, European Journal of Personality, 15, 3-18.

[16]See www.visualdna.com/financial-solutions

[17]Leutner F, Yearsley A, Codrenu S-C, Borenstein Y, and Ahmetoglu, G, (2017), From Likert scales to images: Validating a novel creativity measure with image based response scales, Personality and Individual differences, 106, 36-40

[18]Personality ‘prototypes’ in the Neopic™ process are developed further in the Neo-Archetype engine and linked via Jung’s theory of universal archetypes to predict the energy levels, drive and motivations of respondents.

[19]Reliability refers to consistency in the measurement (numerical score) recorded. Validity refers to the relationship that the measure has to ‘real-world’ contexts and environments e.g. does a high score on ‘extraversion’ predict good leadership?

[20]In widening the scope of the NEOPIC © process, we have also developed a tool that broadens the application of this unique use of imagery. This engine measures personality ‘prototypes’ and then links these to the universal archetypal classification developed by Jung. The Archetypes produced are based upon the values currently held by individuals and are therefore predictive of the energy, drive and motivations of respondents in particular environments and circumstances.

Author: Innovationbubble