=Paper=
{{Paper
|id=Vol-1680/paper7
|storemode=property
|title=A Jungian based framework for Artificial Personality Synthesis
|pdfUrl=https://ceur-ws.org/Vol-1680/paper7.pdf
|volume=Vol-1680
|authors=David Mascarenas
|dblpUrl=https://dblp.org/rec/conf/recsys/Mascarenas16
}}
==A Jungian based framework for Artificial Personality Synthesis==
https://ceur-ws.org/Vol-1680/paper7.pdf
A Jungian based framework for Artificial Personality
Synthesis
David Mascarenas
Los Alamos National Laboratory – Engineering Institute
PO Box 1663 MS T001
Los Alamos, NM, 87544
1-505-665-0881
dmascarenas@lanl.gov
ABSTRACT this article adopt personality traits (the BF in 76 cases out of 81)
The field of computational intelligence has enjoyed much success and, to the best of our knowledge, no other theories were ever
in developing a variety of algorithms that emulate human adopted in computing oriented research. On one hand, this barely
cognition. However, a framework to tie these algorithms together reflects the dominant position of trait based models in personality
in a coherent manner to create a machines that possess the full psychology. On the other hand, trait-based models represent
spectrum of human-like personalities is still needed. To date, personality in terms of numerical values, a form particularly
research on artificial personality synthesis has focused on using suitable for computer processing. [1].” This statement raises an
the Big Five model from the field of personality psychology. The interesting question. Why has computational personality research
overlooked Achilles heel of Big Five (BF) is that it is purely data- thus far restricted itself to exploring trait-based models of
driven model, and thus offers only marginal guidance on how a personality? Vinciarelli points out a number of competing
machine with a personality might actually be created. In this work personality models including: psychoanalytic, cognitive, and
an alternative computational personality framework is presented behaviorist and biological. Arguably, personality models such as
based on the work of Carl Jung. There are two key insights that the biological model may currently not be adequately understood
suggest a Jungian type-based framework is suitable for to be implemented on a computer, and as Vinciarelli points out,
synthesizing an artificial personality. First, the cognitive functions the numerical nature of trait-based models may be amenable to
which form the building blocks of the Jungian personality model computer implementation to a degree. However, the nature of the
can be mapped to classes of algorithms used to emulate cognition. Jungian type-based model also lends itself to implementation on a
Second, the Jungian framework suggests that at any given time computer, but to the best knowledge of the author has not been
humans are only using one of the cognitive functions. This explored to date. This work will outline how Jungian
suggests that a human personality could be emulated using a state psychological type theory can be used as guidance to synthesize
machine with each state implemented using the appropriate class an artificial personality.
of algorithms. This Jungian type-based framework possesses a number of
attractive properties. First, a very large fraction of algorithms
CCS Concepts emulating cognitive processes can be utilized by the framework in
• Human-centered computing➝Interaction design theory, a coherent manner. Second, the framework can arguably
concepts and paradigms synthesize the full spectrum of human personalities. Third,
Jungian-based personality theory is very popular among
Keywords laypeople. Modern personality psychologists might argue that this
artificial personality synthesis; Carl Jung; affective computing. is due to the cognitive bias known as the Barnum effect. From a
Turing test perspective however this is irrelevant, and perhaps
even an advantage because it would facilitate the illusion of a
1. INTRODUCTION machine having a human-like personality For practical
The advent of ubiquitous computing has increased interest in applications all that really matters is that the machine can
techniques for endowing a machine with a human-like convince a human that it has a personality.
personality. Vinciarelli [1] provides an overview of personality
computing research which shows that since 2006 a marked 2. THE JUNGIAN PERSONALITY TYPE
increase in research papers with the word “personality,” included MODEL
in the title. Vinciarelli makes one particularly interesting Carl Jung is probably most popularly known for introducing
statement with regards to the nature of the work that has been the concept of the “introvert,” and “extrovert.” Jung detailed
undertaken thus far, “Trait based models are widely accepted in his theory of psychological types in 1921 [2], [3]. Jung’s
the computing community as well. All of the works surveyed in model for personality is based on the idea of “cognitive
functions.” Jung identified two fundamentally different kinds
of cognitive functions known as “perception,” and
“judgement. [4]” Perception describes how a person takes in
information, and judgment pertains to decision-making. Jung
then broke these classes down further. Jung asserted that
perception came in two main forms: “sensation” and
EMPIRE 2016, September 16, 2016, Boston, MA, USA. “intuition.” Sensation is focused on physical reality:. It tends
Copyright held by the author(s). to focus on the present and past. Intuition is primarily focused
on finding meaning, patterns, or possibilities in information.
The tendency is to focus more on the future [5]. Likewise, were allowed to assume. One of the constraints Meyers &
Jung identified two forms of judgement which he referred to Briggs introduced was that the hierarchy of cognitive
as “thinking,” and “feeling.” Thinking refers to decision- functions had to alternate between introverted and extroverted
making processes that focus more on the application of basic orientations. It is worth noting that the writing of Carl Jung
truths/principles. It tends to impersonal in the sense that it can be interpreted to indicate a different scheme for ordering
resists allowing personal value judgements, or the value the cognitive functions, and competing ordering systems are
judgements of others influence decision making. Conversely, in existence. For the sake of clarity, the author tends to prefer
“Feeling,” puts significant weight on values. These values can the ordering system outlined in [4]. However, with regards to
be either personal or shared by a community. It tends to prefer artificial personality synthesis, the framework outlined in this
decisions that will result in harmony [6]. Jung then further work is flexible enough to be adapted to any desired cognitive
refined his cognitive functions by asserting that each of the function ordering scheme.
four functions has an introverted and extroverted orientation.
An introverted orientation implies a tendency towards a Ultimately, based on their imposed constraints on cognitive
person’s interior world of thoughts, ideas, feelings and function order, Meyers & Briggs identified 16 unique
memories. An extroverted orientation focuses on people or personality types. These personality types were given four-
experiences external to the self [4]. Jung’s clinical letter labels. The first letter is either E or I to indicate an
observations and reflection ultimately resulted in a total of 8 introverted or extraverted orientation of the dominant
cognitive functions. For completeness the eight cognitive cognitive function. The second letter is either S or N to
functions are: indicate the dominant perception preference of sensation or
intuition respectively, the third label is T or F to indicate the
Extroverted Thinking (Te) preferred judging cognitive function of either thinking or
Introverted Thinking (Ti) feeling, and the last letter is either J or P to indicate whether
Extroverted Feeling (Fe) or not the preferred perception function has an introverted
Introverted Feeling (Fi) orientation (J) or an extroverted orientation (P). An example
Extroverted Sensing (Se) of a personality label from the use of this model would be
Introverted Sensing (Si) ENTJ. This would indicate a personality who’s dominant
Extroverted Intuition (Ne) function is extroverted thinking with introverted intuition as
Introverted Intuition (Ni) the auxiliary perceiving function. The constraints imposed by
Meyers & Briggs on the cognitive function order would then
For the sake of brevity, a full description of Carl Jung’s 8 further specify that the tertiary function is extroverted sensing
cognitive functions will not be provided in this document. and the inferior function is introverted feeling.
Since Jung initially introduced the concept of cognitive
functions, the language used to describe them has evolved as There is a very widely held misconception that each of the
has an understanding of their nature. For the purpose of this four letters indicates a dimension of personality, like that
work, the cognitive function descriptions provided in [4] will found in the Big-5 model. It cannot be stressed enough that
be used throughout this work. the four letters used to provide a personality label are in no
way representative of dimensions. It is more appropriate the
In 1923, Katherine Briggs and her daughter Isabel Briggs think of each grouping of four letters as a label. The Jungian
Meyers were exposed to the newly available English model is not based on the concept of dimensions in any way.
translation of “Psychological Types” [7], [8] At the time It is based on the concept of cognitive functions and the
Katherine was in the process of developing her own hierarchical preference with which people with different types
personality theory motivated partially by a desire to of personality use them. Another common misconception is
understand the personality of her son-in-law. Upon reading that the Jungian/Meyers & Briggs model implies a binary
Jung’s work they came to the conclusion that Jung’s theory distinction between -for instance- thinkers and feelers, or
was superior to their own and so decided to adopt and refine judgers and perceivers. The common criticism is that the
the Jungian model. Over the next 20 years the mother- model implies that a person solely uses only one class of
daughter team obsessively observed and documented human cognitive function or the other. For example, that a person is
nature with regards to type. They eventually created a either a thinker or a feeler. Once again this is not how the
variation on Jung’s system with an associated sorting theory works. The theory indicates that all people have access
instrument known as the Myers-Briggs Type Indicator to use all the cognitive functions. It is just that people have
(MBTI). different orders of preference for different functions.
Alternatively, Berens and Nardi [4] explain the preference in
The evolution of Jung’s psychological type proposed by terms of energy expenditure. They describe the use of a given
Myers and Briggs had a few important characteristics that are cognitive function as requiring the expenditure of more or
worth mentioning. First, Myers and Briggs observed that all less energy depending on a person’s personality type.
people used all of the cognitive functions. The dissimilarity Thinking of cognitive function use in terms of energy usage is
between different types of people was the preference with a very convenient way to guide the selection of cognitive
which they used the cognitive functions. In this model each function to use in a given situation because it interfaces well
person uses the cognitive functions in a hierarchical manner with computational thinking on cost functions in optimization
with a dominant cognitive function, followed by an auxiliary, as well as with results in psychology that suggest that self-
tertiary and inferior cognitive functions. Classically, Meyer & regulation relies on glucose/energy levels [9].
Briggs focused on the first four cognitive functions.
Furthermore, Myers & Briggs introduced very specific
constraints on the hierarchical order the cognitive functions
33. A FRAM
MEWORK K FOR MAP
PPING is less common for the ENT TJ machine to use extraverrted
sennsing (Se) and it rarely if ev er transitions tto the introverrted
JJUNGIAN COGNITIIVE FUNCTIONS TO
O feeeling (Fi) statee. Personality is emulated bby executing the
AALGORIT THMS FOR
R EMULATTING alggorithms that c orrespond to thhe current cognnitive functionn.
PPERSONA ALITY
TThe proposed Jungian typee-based frameework for artiificial It is worth not ing that alterrnatives to th e Markov ch ain
ppersonality synnthesis is basedd on three key principles. reppresentation coould be envisiioned. For ins tance, instead of
1. Algorrithms that havve been develooped to emulatee proobabilities, thee proclivity to transition to another
a cognit ive
cogniition (e.g. Prinnciple Compon ents Analysis [10],
[ funnction could bee couched in teerms of energyy costs associaated
Artifi
ficial Neural Neetworks, Lineaar Classifiers, eetc.) witth the use of a cognitive fuunction for a given
g personallity
can be
b mapped to thhe cognitive fuunctions that m make typpe as discuss ed in [4]. A lso, please nnote that onlyy 4
up thhe building bloccks of the Junggian personalit y coggnitive functioons are shown in Figure 1. F For completen ess
modeel. These algorithms can be uused to implem ment all 8 cognitive functions
f couldd be included. Only the firsst 4
the coognitive functiions. funnctions were used
u as is speccified by the c lassic Myers aand
2. Humaan personality is inherently sserial in naturee. Briiggs framewoork for the ssake of claritty. Lower ordder
Humaan personality arises from th e limitation that funnctions were not included. Inn initial implem mentations of tthis
humaans can only usse one cognitivve function at aany fraamework it m might be prefeerable to onlyy use the firstt 2
givenn time. Or at leeast the human ability to use more higghest preferennce cognitive functions fo for the sake of
than one function a t a time appearrs to be severe ly simmplicity. The reason being that the firsst two cognit ive
limiteed. funnctions are all that is neededd to distinguishh one personallity
3. The JJungian type-b ased personali ty framework typpe from the othher 16. Furthe rmore Jungianns tend to belieeve
allow
ws for the possiibility of an inddividual agent to thee first two are the
t most imporrtant.
use aany of the 8 coggnitive functioons at any givenn
time, however, perssonality emerges from a In order to impplement the J ungian type-bbased personallity
hiera rchical prefereence for certainn cognitive fraamework, it iss necessary too map algorithhms that emullate
functtions over othe rs. The order aand magnitude of coggnitive processses to the ccognitive funcctions. Table 1.
prefeerence can be s elected based oon models suchh as proovides a listi ng of the 8 cognitive fu nctions, a shhort
the M
Myer & Briggs cognitive funcction orders [4]]. desscription of eaach of the coggnitive functioons, and a list of
Theyy could also posssibly be learnned from data, oor po ssible algorithhms that couuld be used to emulate the
couldd even be choseen arbitrarily. IInitially the auuthors coggnitive functioon. This list iss by no meanss exhaustive, aand
suggeest using the MMyers & Briggss cognitive funnction revvisions are expected
e as artificial
a pers onality researrch
orderrs as guidance. proogresses. The descriptions provided in T Table 1. borrrow
heaavily from [4 ] in order to maintain som me uniformity in
desscriptions.
T
Table 1: Cogniitive functionss and associatted algorithmiic
caandidates for iimplementatioon.
C
Cognitive D
Description [4]], Candidatte Algorithms
F
Function
Partiaally Observablee
Exxtroverted O
Organizing Markoov Decision
T
Thinking peeople and Proceess
(Te) thhings to work
A*
effficiently and
prroductively. Optimmization
O
Organizes the routinnes
ennvironment andd Lineaar
iddeas. Progrramming
Princiiple
Inntroverted A
An internal Compponents
T
Thinking seense of the Analyysis
(Ti) esssential
Indeppendent
quualities of
Compponents
soomething.
Analyysis
F
Figure 1 A Markov chaain representaative of an E ENTJ. N
Noticing the fin e
diistinctions thatt Spars e Dictionary
C
Cognitive funcction preferencce order: Te – N
Ni – Se – Fi.
m
make it what it Learnning
FFigure 1 show s an implemenntation of the Jungian type-bbased is , internal Auto Associative
ppersonality fraamework in thhe form of a M Markov chain . The reeasoning Neuraal Networks
sstates of the Markov chaain are occuppied by cognnitive prrocess of
ffunctions. Thee values assocciated with thhe edges prov ide a deeriving
pprobability th at the artificiial personalityy will transitioon to suubcategories off
eeach of the allternative cognnitive functionns. In the casee of a suub-classes and
mmachine with an ENTJ persoonality, a procclivity will exi st for suub-principles oof
tthe machine too remain in a n extroverted thinking (Te) state geeneral
wwith occasionaal transitioningg to the intuitivve intuition staate. It prrinciples.
Use own alternative
Extroverted The desire to embodiment as subject.
Feeling (Fe) connect/disconn analog computation Supervised learning
ect with others. based on perception Introverted Storing
Causes response Support Vector
of external affect Sensing (Si) experiences and
to expressed or Machine
information and
Artificial Neural Classification
unexpressed comparing/contr
Networks Matched filtering
needs of others. asting the
Takes on the Resources allocated Narrowband
current situation
feelings of weighted towards filtering
with similar
others – communications/col Autocorrelation
prior
Empathy laboration with Cross correlation
experiences.
other agents
The
(human, machine
similarities/diffe
and otherwise)
rences are
State of health registered as
evaluation of other important input.
agents
Search Engines
Cost functions for Extraverted Cross- (web)
optimization Intuition contextual,
Compressive
designed in such a (Ne) divergent
Sampling
way that rewards thinking.
associated with Generates and Random Walk
group success explores a host Genetic Algorithms
outweigh individual of possible
rewards associated interpretations
with individual from a single
success idea. The ability
Analysis of how to entertain a
actions will affect variety of
group well-being disparate ideas,
beliefs and
meanings
Artificial Neural simultaneously
Introverted A filter for Networks
Feelings information that while
Techniques for state maintaining the
(Fi) matches what is
of health monitoring possibility that
valued, wanted,
of self. they are all true.
or worth
believing in. Cost functions for Seeing things
Continual optimization “as if.”
assessment of a designed in such a Simulation/Predicti
way that individual Introverted Lays out how on
given situations Intuition the future might
with respect to rewards for Design of
individual success (Ni) unfold based on
individual Experiment
outweigh group unseen trends
values. and signs. Can Autocomplete
rewards associated System ID
with group success involve working
out complex Interpolation/Extrap
Analysis of how olation
actions will affect concepts or
systems of Bayesian Inference
individual well-
being thinking or
Active learning conceiving of
Extroverted Use of the symbolic or
Active SLAM
Sensing concrete senses novel ways to
(Se) to become aware Online Learning understand
of the physical Search based on things that are
world in detail. maximum universal.
An impulse to information gain
act on PID control The introverted and extroverted feeling judging functions
information in merit additional discussion. The feeling judging functions are
order to get often associated with emotional response. The author
immediate currently likes to think of emotional decision-making in
results. Active general to be similar to an artificial neural network in the
seeking of
sense that an artificial neural network can often take many
information until
complicated inputs and learn relationships between the inputs
sources of input
are exhausted or that can be used to quickly make decisions, however it is not
attention is always clear how exactly those decisions are made. The
captured by author also prefers to think of the emotional cognitive
functions as being partially the results of an embodied of two groups with respect to each letter in the Meyers-Briggs
intelligence. Feeling judgements can be thought of as using labels. Some research suggests this bimodal distribution was
the body as an analog computer to perform simulations and an artifact of the analysis techniques used [15]. The problem
make decisions. Work by Nummenmaa [10] on mapping the with the original argument was that it was not necessarily
sensation of emotions felt in the human body lends some respecting the Jungian model as a dynamic system and was
support for this perspective. making the assumption that the Jungian model could be
represented with a four dimensional Cartesian coordinate
As an example of the application of the framework, a machine system. It is not clear what type of distribution a personality
endowed with an ENTJ personality using the Markov chain in consisting of cognitive functions as building blocks should
Figure 1, we might employ the A* algorithm to perform the generate when evaluated using a self-report inventory. A
dominant extroverted thinking (Te) cognitive function to form central limit theorem argument could be made to suggest it
a plan to move through an environment. After the plans are come out as a Gaussian, but to really make a definitive
generated they might be analyzed in the introverted intuition statement a more rigorous analysis should be undertaken.
(Ni) state using an appropriate simulation that corresponds to Ultimately when synthesizing an artificial personality, the
the environment and task of interest. If the simulation most important criteria for most applications is that the
verifies that the plan is acceptable the plan may be executed personality be convincing to humans. As long as the machine
by the extroverted Sensing function. In many cases the can pass a Turing-like test it is probably an acceptable
machine may totally ignore the introverted feeling state and solution. The fact that the Jungian model is so widely popular
proceed directly to execution, but when it does go into the suggests that it may have a low barrier to acceptance among
introverted feeling state it may use a neural network trained to the majority of the population.
look out for the machine’s own well-being to decide whether
or not to actually execute the task based on whether or not the
machine “feels,” it will advance the machine’s self-interests. 5. JUSTIFICATION FROM A
NEUROSCIENCE PERSPECTIVE
It is also worth noting that creating the ability to endow A number of interesting results have come out of the
robots with personalities may also have application in human psychology and neuroscience communities that lends some
machine teaming applications. Much as diversity in teams of support for the idea of using the Jungian type model as
humans tends to lead to better results [4], endowing teams of outlined in this work. The idea that humans can effectively
robots with different personalities may also lead to more only use one cognitive function at a time gains support from
robust robotic teaming. For example, a team consisting of the results of Watson [16] that suggest 98% of humans are
humans could be augmented with machines endowed with incapable of multitasking. Jack’s analysis of fMRI
personalities different from the existing team members in measurements of the human brain suggested that there are
order to enhance the overall team diversity. Teams consisting physiological constraints on our ability to simultaneously
only of robots could be designed in such a way that individual engage two distinct cognitive modes. In this case they found
team members are endowed with complimentary humans could not attend to tasks that require social cognition
personalities, thus enhancing the overall robustness and and physical reasoning simultaneously [17]. Grondin (2015)
performance of the machine team. [18] found neuroanatomical differences between Agentic
(achievement-oriented) and Affiliative (sociable) extroverts.
4. ADDRESSING SOME CONCERNS From a Jungian perspective the concept of an agentic
WITH THE JUNGIAN MODEL extroverts corresponds very well with personality types with a
Since the MBTI was developed more than 10,000 companies, dominant extroverted thinking preference (ENTJ, ESTJ) and
2500 colleges/universities and 200 U.S. government agencies affiliative extroverts correspond strongly with personality
have used the test [11]. It is estimated that more the 50 types with a dominant extroverted feeling preference (ENFJ,
million people have used the instrument since 1962. Jungian ESFJ).
personality theory has had a great influence on corporate
America as well as popular culture. For instance, Carl Jung 6. CONCLUSIONS
popularized the terms introvert and extrovert. Despite the
Personality from a Jungian perspective can be thought of as a
popularity of Jungian personality theory in industry and zero sum game. Humans only have limited cognitive
among laypeople, academia tends to discount it. One common resources, and our personality is based on how we tend to
criticism against the MBTI is that it lacks test-retest choose to use those resources. This proposed framework is
reliability [12]. The current perspective of the author is that particularly attractive because it uses established algorithms
this criticism is probably valid. At this time the instrument as building blocks for personality. Because the building
itself appears to have problems. The reason for the lack of blocks are algorithms, and in some cases learning algorithms,
reliability with the current instruments may be that the the machine is ultimately able to learn and adapt to
instrument is based on the analysis of a self-report inventory. experiences. The personality framework provides a genotype
This type of instrument may be suitable if a Cartesian model so to speak but the ultimate phenotype of the machine
of personality is used, but the Jungian model is better depends on the experiences it encounters throughout its span
described as a dynamic system. Ultimately enhanced versions of existence. This framework allows for great diversity in
of projective tests such as those suggested by Ottley [13], and resulting perceived personality phenotype.
Brown’s [14] work may be better able to characterize human An interesting implication of the Jungian type-based
personality. Another common criticism is that some Jungian personality framework is that it might help in the
advocates made the claim that MBTI score distributions development of robust, high-performance human-machine
assumed bimodal distributions, thus lending support to the teams consisting of members with diverse personalities.
misunderstanding of Jungian theory that people fell into one Observations on the personality diversity of teams and their
performance suggest that teams consisting of members with Others: An Introduction to the Personality Type Code,
diverse personalities tend to perform better [19]. A team Huntington Beach, CA: Telos Publications, August 1, 2004.
consisting of humans could be augmented by machines
endowed with personalities that enhance the team’s overall [5] The Myers and Briggs Foundation, [Online]. Available:
personality diversity. Alternatively, teams consisting solely of http://www.myersbriggs.org/my-mbti-personality-type/mbti-
machines could consist of members who provide each other basics/sensing-or-intuition.htm. [Accessed 1 June 2015].
with different perspectives of the world they are interacting
with. [6] The Myers & Briggs Foundation, [Online]. Available:
Experience has shown that data science problems often benefit http://www.myersbriggs.org/my-mbti-personality-type/mbti-
from the use of a combination of many heterogeneous models. basics/thinking-or-feeling.htm. [Accessed 1 June 2015].
The Netflix prize provides a good example of showing the [7] The Meyers & Briggs Foundation, [Online]. Available:
advantages associated with simply combining different http://www.myersbriggs.org/my-mbti-personality-type/mbti-
approaches [20]. However contemporary computational basics/isabel-briggs-myers.htm. [Accessed 1 June 2015].
personality research is dangerously homogeneous in the sense that
all computational personality research is currently using a trait- [8] Center for Applications of Psychological Types, [Online].
based paradigm [1]. The field would benefit from competing Available: http://www.capt.org/mbti-assessment/isabel-
models and approaches. Furthermore, from an engineering/Turing myers.htm. [Accessed 1 June 2015].
test perspective it really does not matter whether or not the
approach used to generate an artificial personality accurately [9] M. Gailliot, R. Baumeister, C. DeWall, J. Maner, E. Plant, D.
models what is occurring in the human mind. It is only necessary Tice, L. Brewer and B. Schmeichel, "Self-control relies on
to convince the person interacting with it that it is a human-like glucose as a limited energy source: willpower is more than a
personality. In fact, the widespread popularity of Meyer’s Briggs metaphor," J Pers Soc Psychol, vol. 92, no. 2, pp. 325-336,
in business settings and with the general public suggests that an February 2007.
artificial personality based on the Jungian model may perform [10] K. Pearson, "On Lines and Planes of Closest Fit to Systems
well with respect to convincing other people of the machine’s of Points in Space," Philosophical Magazine, vol. 2, no. 11,
personality. pp. 559-572, 1901.
Artificial personality synthesis is at a similar point in [11] L. Nummenmaa, E. Glerean, R. Hari and J. K. Hietanen,
development as the airplane was at the time of the Wright
"Bodily maps of emotions," Proceedings of the National
Brothers. You do not need to understand how a bird flies in
order to build a transatlantic aircraft, and you do not need to Academy of Sciences of the United States of America, vol.
understand how the human mind works in order to build a 111, no. 2, pp. 646-651, November 27 2013.
machine that has a recognizable personality. Finally, it is [12] L. Cunningham, April 13, 2013. [Online]. Available:
interesting to note that while the Jungian perspective on
http://www.seattletimes.com/business/myers-briggs-
artificial personality synthesis presented here is very different
from the mainstream views on the topic that center on the use personality-test-embraced-by-employers-not-all-
of Big-5 theory, this paradigm is not any more controversial psychologists/. [Accessed 1 June 2015].
than the Freudian paradigm advocated for by Marvin Minsky [13] D. J. Pittenger, "Measuring the MBTI... And Coming Up
[21]. The Jungian paradigm presented in this work is worth Short," Journal of Career Planning and Employment, vol.
consideration in future artificial personality synthesis. It is
54, no. 1, pp. 48-52, 1993.
attractive in the sense that it can leverage a variety of existing
algorithms to implement the personality. It also has the [14] A. Ottley, H. Yang and R. Chang, "Personality as a Predictor
property that an artificial personality made with this of User Strategy: How Locus of Control Affects Search
framework should exhibit the full range of human personality Strategies on Tree Visualizations," in ACM SIGCHI
as described by Jung.
Conference on Human Factors in Computing Systems (CHI),
7. ACKNOWLEDGMENTS Seoul, Korea, 2015.
This work was completed under the support of a Los Alamos [15] E. Brown, A. Ottley, J. Zhao, Q. Lin, A. Endert, R. Souvenir
National Laboratory Early Career Research Award and R. Chang, "Finding Waldo: Learning about Users from
#20140629ECR. The concepts presented in this work were their Interactions," IEEE Transactions on Visualization and
refined with the help of Sebastián A. Zanlongo. Computer Graphics (TVCG), vol. 20, no. 14, December
2014.
8. REFERENCES
[16] T. Bess and H. R.J., "Bimodal score distributions and the
Meyers-Briggs Type Indicator: Fact or Artifact," J Pers
[1] A. Vinciarelli and G. Mohammadi, "A Survey of Persoanlity Assess, vol. 78, no. 1, pp. 176-186, February 2002.
Computing," IEEE Transactions on Affective Computing,
vol. 5, no. 3, pp. 273-291, June 2014. [17] J. Watson and D. Strayer, "Supertaskers: Profiles in
extraordinary multitasking ability," Psychonomic Bulletin &
[2] C. G. Jung, Psychologische Typen (Psychological Types), 1 Review, vol. 17, no. 4, pp. 479-485, 2010.
ed., Zurich: Rascher Verlag, 1921.
[18] A. I. Jack, A. Dawson, K. Begany, R. L. Leckie, K. Barry, A.
[3] C. Jung, The Portable Jung, J. Cambell, Ed., New York: The Ciccia and A. Snyder, "fMRI reveals reciprocal inhibition
Viking Press, 1976. between social and physical cognitive domains,"
NeuroImage, vol. 0, pp. 385-401, 2012.
[4] L. V. Berens and D. Nardi, Understanding Yourself and
[19] E. Grodin and T. L. White, "The neuroanatomical delineation
of agentic and affiliative extraversion," Cognitive, Affective,
& Behavioral Neuroscience, vol. 15, no. 2, pp. 321-334, June
2015.
[20] D. Wilde, "More Diverse Personalities Mean More
Successful Teams," ASME, March 2011.
[21] E. V. Buskirk, "How the Netflix Prize was Won," Wired, 22
Sept 2009.
[22] M. Minsky, The Emotion Machine, Simon & Schuster, 2006.