- The era of User Generated Content (USG) on Social Networks has invested different areas of human experience, including one of the highest levels of education, i.e. the PhD grade. As regards the opportunity to "imitate" a virtual content, the phenomenon of Memes (Shifman, 2014) is spreading. The research has two goals: to demonstrate how the practices of signification of humor on the doctorate are modified with the "memes factory" and how artificial intelligence is able to detect a complex rhetorical strategy such as humor. To satisfy these purposes, 39 memes about the PhD were collected and, then, analyzed in two levels: the quanti-qualitative, aiming to detect the semi-automatic emotional involvement, expression of humor in “meme discourses”; the semiotic analysis of meme images to understand if it is possible to classify PhD humor. The analysis show a further function of humor mediated by communication through memes, that is the complaint against a complex and precarious career path.
Keywords—PhD Memes, Mediated Humor, Emotional Analysis, Diatextual Analysis Semiotic, Analysis of Imagines
I.
INTRODUCTION
The era of User Generated Content (USG) on Social
Networks has invested different areas of human experience,
including one of the highest levels of education, i.e. the PhD
grade. In fact, Social Networks, like Facebook, amplify the
possibility of creating content and making sure that they reach
a wide audience. Indeed, every user, even a non-expert, can put
content online, share it or imitate it. As regards the opportunity
to "imitate" a virtual content, the phenomenon of Memes is
spreading. Conceptually the Meme is "the minimum cultural
unit capable of replication"
Taking the theoretical boundary between memetics, psychology and computer science, the study aims to assess how, in a context mediated like Facebook, the Memes convey the humorous content regarding a highly qualified educational path, such as the PhD, which is a high level of education, but also delicate because PhD students are neither a teachers/workers nor a students. Furthermore, the research has two goals: to demonstrate how the practices of signification of humor on the doctorate are modified with the "memes factory" and how artificial intelligence is able to detect a complex rhetorical strategy such as humor.
To satisfy these purposes, 39 memes about the PhD were
collected from a Facebook page called “High Impact PhD
meme” during the first part of 2019. The PhD memes were
analyzed in two levels: the quanti-qualitative, aiming to detect
the semi-automatic emotional involvement, expression of
humor in “meme discourses”; the semiotic analysis of meme
images to understand if it is possible to classify PhD humor.
The emotional involvement were detected by means of Tone
Analyzer IBM
Regarding the visual part of a meme, a semiotic analysis of images was carried out, in which the main features were highlighted in terms of structure, colors and principal subjects.
II.
INTELLIGENCE
The psychological interest in humor establishes roots in the
psychoanalytic theories
Following these theories, people laugh at things that are
unexpected or surprising and it is the violation of an expected
pattern that provokes humor in the mind of the receiver. Rather
than focusing on the physiological (relief theory) or emotional
(superiority theory) function of humor, incongruity theory
emphasizes cognition. It assumes that the cognitive capacity to
note and understand incongruous events is necessary to
experience laughter or mirth. Absurdity, nonsense, and surprise
are vital themes in humor covered by this theory
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). techniques fall into four general categories: language, logic, identity, and action. Some examples are summarized in Table I:
This theoretical framework begins to be challenged
following the Digital Revolution and the introduction of new
media, in which it has been experienced a humorous situation in
a different way. In particular, the spread of social networks,
which offer the possibility of producing a quantity of content
very easily, modifies the experience of the humorous
communication situation, the function and the comprehension
itself of this rhetorical strategy. First of all, the main difficulty
regarding the mediated humor is the absence of enunciators face
to face interaction. In fact, this limit engrave the possibility of a
fast understanding, such as in the case of mediated irony
Identify applicable funding agency here. If none, delete this text box. that users that know the enunciation context may understand easily the meaning of communication act, on the contrary, there was no effect of joy and mirth in the humorous meaning.
Currently, the studies about Artificial Intelligence (AI) and Natural Language Processing (NLP) focused on the creation of sophisticated software, able to detect and recognize the tone and the emotions of all discourse genres. It is the case of IBM tools, in particular the Tone Analyzer, that is more than an Artificial Intelligence, because it may be classified as Affective Intelligence. The mode of operation of these software is that of matching the text in input with their inner vocabularies of words, going back, as output, a data of total emotion on the base of a scale with score lower than 0,50, equal to 0,50 and higher than 0,50: the significant values are those higher than 0.50. The emotions detected are the five basic emotions, which are: joy, fear, sadness, disgust and anger. Joy or happiness has shades of enjoyment, satisfaction and pleasure. There is a sense of wellbeing, inner peace, love, safety and contentment; while fear is a response to impending danger. It is a survival mechanism that is a reaction to some negative stimulus. It may be a mild caution or an extreme phobia; sadness indicates a feeling of loss and disadvantage. When a person can be observed to be quiet, less energetic and withdrawn, it may be inferred that sadness exists; disgust is an emotional response of revulsion to something considered offensive or unpleasant. It is a sensation that refers to something revolting and anger evoked due to injustice, conflict, humiliation, negligence or betrayal. If anger is active, the individual attacks the target, verbally or physically. If anger is passive, the person silently sulks and feels tension and hostility. But, even in these cases, the main limit is that it is impossible reduce the complexity of human experience in a matching: in particular the AI is not able to detect all rethorical strategies and all context.
who invest three or four years in the in-depth study of specific scientific topics. These are graduates with different educational backgrounds, from engineers, doctors to more humanistic disciplines such as social sciences, philosophy and literature. They are, however, people who, after completing their university studies, transform their "thirst for knowledge" into an occupation.
However, it is an occupation that is more "vulnerable" to the labor market flexibility logics, in a context, such as the academic one and, on a larger scale, the scientific one, which is based on rigor and quantity. In addition to this, there is a lack of clarification in the definition of who, in terms of skills and tasks, is a "doctoral student". In most cases, it is a paid "qualified" student, but not all PhDs are eligible for financial support. The common element is that all those who undertake this career clash with an uphill path, not always linear and characterized by deadlines, pressures, measurements of their work through purely quantitative indices, all surrounded by the uncertainty of the future.
This landscape has aroused the interest of sociologists, psychologists and economists, who, starting from statistics, have begun to study doctoral students, with "research in research". In particular, a Belgian research of 2017, published in Research Policy, studied the relationship between the academic career and the mental health of the doctoral students (Levecquea et al., 2017). Specifically, 3600 Belgian doctoral students of all disciplines were involved and it emerged that, independently of training, one third of the sample had a high risk of developing, above all, depression. Despite this, there are no other studies to confirm the data, as most of the results show a correlation with stress and states of anxiety, mainly identified as a concern for the future, due to the lack of prospects. There is also the absence, in the universities themselves, of professionals who deal with supporting Human Resources in future employment or simply accepting individual and collective difficulties and needs. Some strategies, already present within the Social Networks, useful to defuse the common practices of meaning related to the doctorate, are the international humorous pages, in which every PhD student is recognized. These pages, like many virtual communities, have almost a "mutual help" function, because, thanks to sharing, it is possible to feel less alone.
This is the case of a specific communication tool, which is becoming a rapidly spreading cultural artifact, i.e. the "Meme". The meme is composed of two communicative elements: one visual and the other textual.
The research has two goals: to demonstrate how the practices of signification of humor on the doctorate are modified with the "memes factory" and how artificial intelligence is able to detect a complex rhetorical strategy such as humor. To satisfy these purposes, in the study, 39 memes about the PhD were collected from a Facebook page called “High Impact PhD meme” during the first part of 2019. In fact, more recent PhD memes have been chosen. The PhD memes were analyzed in two levels: the quanti-qualitative, aiming to detect the semiautomatic emotional involvement, expression of humor in “meme discourses”; the semiotic analysis of meme images to understand if it is possible to classify PhD humor.
From the Emotional Analysis of the PhD memes, a significant prevalence of neutrality and negative emotions such as anger, sadness or fear emerges, as can be seen in chart 1:
Chart 1: Quantitative results of Emotional Analysis By convention, following the model of Barrett and Russell (2009) which disfigures positive emotions from negative emotions, the results of positive emotions (joy) have been given a value of 1; for negative emotions (anger, sadness, disgust and fear) a value equal to -1 and for neutrality a value equal to zero. The results show the prevalence of neutrality, represented in gray in the graph. The negative emotions in red are also significant, where, however, the absence of disgust is highlighted. The positive emotions represented in green are quantitatively less significant. This is an important indicator because, from the psycholinguistic scientific literature on “humor”, the connection of this rhetorical strategy to a positive emotion emerges. Starting from this limit and considering the memes discourses as “diatexts”, whose meaning derives from the interweaving of text and context, it can be noted that the emotionality detected by the Artificial Intelligence is not entirely incorrect, as in the following example:
revisions. (0,51)
Ah shit, here we go again (0,73)” In this case, the real fear emerges of those who submit a scientific work in the form of an article to a journal and receive it with many corrections to be made, perhaps, in a limited period. Fear is immediately replaced by the anger that emerges in having to work again on the same research, further modifying what could have been the initial idea.
At the textual level, in the memes on the doctorate it can be noted the constant presence of bad words, such as "shit", "bastards" which then load the emotions in a negative sense. On the contrary, in the case of joy detection, this is due to the use of intensifiers, as in the following example: Eg. 2 “When your manuscript is finally revised gets backfrom review” (0,74) Also in this case, the basic theme is the review of scientific articles by journals. The use of the "finally" intensifier loads the sentence of joy.
From these initial results it can be deduced that the PhD practices of significance are the subject of the memes concern the relationship with the supervisors, the difficulties in publishing the papers and the post-doc uncertainty, as shown in these expamples:
People: and then what?
Me: Good question.” In these examples, there is a prevalence of neutrality, linked to a lack of emotions or to the presence of contrasting emotions, i.e. anger and joy in the same phrase. In PhD meme discourses, neutrality is also due to the presence of imagines, that sometimes is the real element of joy. If the image is bigger or more significant than the text, neutrality is justify. Therefore, it is necessary analyze PhD meme images that, in most cases, replace the non verbal communication act.
Sight is one of the most attractive senses of the human
being, as it communicates messages immediately with respect
to the written text. An image is the privileged and mediated way
to pass quickly from the level of the signifier (plan of the
expression) to that of the meaning (plan of the content). The
semiotic analysis of images
After having identified these three characteristics, it can be found the plastic formats that allow to reach the content, that is the transition from the signifier to the meaning, which are divided into two aspects: 1. the symbolic correlation, that is the set of conventions and cultural aspects connected to the expression plan (for example if the gold color indicates sacredness, the subject / object depicted that dresses gold, will be a sacred subject / object); 2. the semisymbolic correlation, which provides a relationship between a category of the expression and one of the content (for example the topology / topology category, can correspond to one of the content where “high” corresponds to “sacred” and “low” to “ profane”.
The application of this methodology aims to understand the presence or absence of regularity on the creation of visual aspect in PhD memes.
First of all, the image becomes the imitation element in the "meme" phenomenon. The images used are actually known. These are images that represent clear and identifiable subjects, placed in the foreground. In some cases, the presence of dialogues is recorded and therefore the image is a vignette, where all the characters have an equal portion of space. Even the colors are clear in most cases, so they can be easily recognized. As for the subjects and the situations portrayed, the protagonists of the photos are famous movie characters with expressions consistent respect to the text of the image. In the case of PhD memes, the recurring expressions in the characters is doubt and sadness, as shown in the following examples: As can be seen from the images, the visual aspect helps the comprehension of the text and completes it. Furthermore, they are well-known TV personalities: in the first case the expression of doubt and the sentence of the protagonist are not original, that is to say that the context in which they are inserted changes, which makes it possible to consider it an imitation and detect the basic humor. In the second case, the expression of original sadness of a well-known protagonist of a TV series for teenagers, is set in a different context, where, however, the image is coherent.
Other characters typical of the meme images on the PhD are the protagonists of cartoons, as in the following examples: The images, however, are not chosen randomly. In example 3 there is an "ogre" in the foreground, which in traditional culture is considered a monstrous figure, here instead replaced by an ogre of modern cinematography used to overturn many of the fairy tales elements. The humor of the PhD meme fits into this reversal.
The era of User Generated Content (USG) on Social Networks has invested different areas of human experience, including one of the highest levels of education, i.e. the PhD grade. As regards the opportunity to "imitate" a virtual content, the phenomenon of Memes is spreading. The research has two goals: to demonstrate how the practices of signification of humor on the doctorate are modified with the "memes factory" and how artificial intelligence is able to detect a complex rhetorical strategy such as humor. To satisfy these purposes, 39 memes about the PhD, analyzed in two levels: the quantiqualitative, aiming to detect the semi-automatic emotional involvement, expression of humor in “meme discourses”; the semiotic analysis of meme images to understand if it is possible to classify PhD humor. The virtual context, but in particular the image, conveys a representation of the academic career and, specifically of the doctoral course, which is often obscured in the offline environment. The virtual context also changes the way we express humorous techniques. Most rhetorical techniques such as absurdity or imitation, with memes precisely, take on a visual form rather than an oral one. In fact, the analysis show a further function of humor mediated by communication through memes: the complaint against a complex and precarious career path.
This reporting function of PhD meme communication act emerges above all from the Emotional Analysis and the prevalence of neutrality and the anger. The visual part is a means to mitigate the discourse meanings. Humor and imitation strategies arise in the contrast between expected/unexpected or known/unknown: generally it is expected that a famous character behave in a certain way, instead it is integrate in a contextual unexpected framework known.
The main features of the PhD memes can be summarized in the following taxonomy: The research shows that humor in the context of PhD memes is based on two levels: a cognitive first linked to the perception of something unexpected and a second emotional level, which explains the presence of negative emotions such as fear. One of the limitations of the research concerns the prevalence of output neutrality. The software does not allow, in case of neutrality, a sentence level analysis able to understand if neutrality is given by the absence of sentiment or as a juxtaposition of positive and negative emotions, an aspect that will be evaluated as a future perspective. In addition, among the future perspectives, a comparison of the current results of the study with other sources and with a larger sample will be provided. An inter-judges agreement will also be possible, in order to avoid the excessive simplification to five emotions carried out by the IBM software.