Written evidence submitted by Loughborough University (SMH0167)

MEMEotive - Analysing the Effects of Internet Memes on Young
Teenagers’ Health and Health Behaviours
Submitted by:
Dr Ash Casey, School of Sport, Exercise and Health Sciences (SSEHS),
Loughborough University
Dr Martin Sykora, School of Business and Economics (SBE), Loughborough
University
Dr Suzanne Elayan, SBE, Loughborough University
Professor Tom Jackson, SBE, Loughborough University
Professor Lorraine Cale, SSEHS, Loughborough University

1. Executive Summary










There is growing concern over the significant costs of unhealthy lifestyles to
the NHS;
This concern is exacerbated by growing evidence of young teenagers’ poor
health and unhealthy lifestyles;
UK youngsters are lagging behind their European counterparts in many areas
(WHO, 2017). For instance, nearly a third of children in England are
overweight or obese (Health and Social Care Information Centre, 2015),
compared to less than a fifth across other European countries (WHO, 2017).
Given the increasing significance of, and reliance on, online sources of health
information it is important to understand the Effects of Internet Memes on
Young Teenagers’ Health and Health Behaviours;
There is evidence of uncritical consumption of on-line health-related
information by teenagers and a lack of regulation and quality control in
relation to on-line health information.
This highlights the critical significance of social media to young teenagers’
development and their wellbeing;
There is a need to be aware of the over optimism of young teenagers with
regard to their own online vulnerability
Our provisional inquiries show that a substantial number of individuals on
Twitter share health related Internet memes, with both positive and negative
messages, through their public accounts. Such is the pervasiveness of
Internet memes that the vast majority of sharers display little, if any, emotion
when sharing these memes: many of which contain inappropriate material or
ridicule others by race, gender, ethnicity, sexuality, body shape, religion, diet
etc. When viewed in this way, Internet memes have the potential to normalise
undesirable behaviours such as trolling, body shaming and bullying, and a
lack of emotion may be indicative of a larger apathy with regards to such
practice.

 Written evidence submitted by Loughborough University (SMH0167)

2. Introduction
1. About us
a. Dr Ash Casey has a strong publication record in education and social
media/technology. His exploration of the role of Internet memes in shaping
teenagers’ health-related social media interactions led to this work. He has 15
years secondary school teaching and 8 years teacher-education experience.
b. Dr Martin Sykora has a strong publishing and funding track record in
computational sentiment analysis, with applications in health and well-being.
He is Loughborough’s PI on the EU Horizon2020 NESTORE project, which
explores computational big-data solutions and technologies for older people to
retain everyday life activities and healthy behavioural decision making.
c. Dr Suzanne Elayan works in semantic modelling and designed and built the
EMOTIVE ontology and other semantic models. She has a successful track
record in ontology-building and design for social media monitoring.
d. Professor Tom Jackson led the EMOTIVE team and since then he, Martin
Sykora and Suzanne Elayan have worked as a team on a series of successful
projects. He has 15 years’ experience of research and industrial consultancy
and has been part of securing over £14.3M from Research Councils.
e. Professor Lorraine Cale has published widely and secured a number of
research grants in the area of health. More recently she has written in the
areas of social media, young people and health and social media. She is
engaged with policy makers and key stakeholders in various capacities, e.g.
she regularly chairs Westminster Policy Briefing events on relevant national
policy developments.
2. Loughborough University is ranked number one in the QS world university
rankings for sport-related subjects (2017 and 2018). In addition, the university has
a very strong international profile in emotion and linguistic based research and
computer system development. These factors combined mean we are strongly
placed to submit evidence to the House of Commons Science and Technology
Committee’s current inquiry on ‘The Impact of Social Media and Screens on
Young People’s Mental Health’.
3. Evidence
3. Unhealthy lifestyles cost the NHS billions every year: smoking reportedly costs
over £5 billion, obesity £4 billion, alcohol £3.5 billion and physical inactivity £1.1
billion (Kings Fund, 2017). These figures are significant given growing concerns
over young teenagers’ health. The prevalence of online health information and the
dangers of inaccurate/inappropriate health messages – coupled with the fact that
searches for health information online have far outpaced searches via other
media sources – mean it is imperative that we study the sources of online health-

 Written evidence submitted by Loughborough University (SMH0167)

related knowledge that young teenagers access. With 13-16 year olds
increasingly seeking to address their own health concerns (Wartella et al. 2015)
MEMEotive will enable us to understand how and why unhealthy behaviour can
become normalised. Investing in and putting young teenagers at the centre of this
work is key to understanding how Internet-memes contribute to the normalisation
of unhealthy lifestyles. At the heart of MEMEotive is the drive to influence the
decisions of policymakers and other stakeholders (e.g. schools) when it comes to
supporting young people to be healthy.
4. Most recent findings from the Health Behaviour in School-aged Children World
Health Organisation Collaborative Cross-National survey reveal worrying patterns
and trends concerning teenagers’ health and health behaviours (including high
obesity levels, low physical activity and poor eating behaviours), with UK
youngsters lagging behind their European counterparts in many areas (World
Health Organization (WHO), 2016). For instance, nearly a third of children in
England are overweight or obese (Health and Social Care Information Centre,
2015), compared to less than a fifth across other European countries (WHO,
2016). Given that health habits and many health problems are established early in
life, the importance of early intervention is crucial (WHO, 2016). Indeed, the
seriousness of the situation is reflected in recent policies aimed at tackling health
issues in young people, i.e. the Childhood Obesity Plan and Towards an Active
Nation. Our research seeks to influence these and other social policy initiatives by
considering hitherto overlooked concerns about Internet health knowledge,
triggered by recognition of the pervasiveness of social media, and the ease of
accessing Internet-based health resources.
5. Young teenagers increasingly search for health information online (Wartella et al.
2015) and concurrently maintain a significant social media presence (Jenkins et
al. 2013). Accompanying the vast reach of social media are concerns about the
potential for the spread of both misinformation and potentially health damaging
Internet memes (Casey, In Press). Given this, and the current trends and patterns
in the health behaviour of young teenagers, it is necessary to better understand
what health knowledge and health messages young teenagers access/acquire on
social media and how they respond. This is especially vital given evidence that
social media is critical to 13-16 year olds’ wellbeing (de Vries et al. 2016) and that
tension exists between young teenagers’ awareness of the risks of social media
and their over optimism with regards to their vulnerability to such risks (Betts and
Spenser, 2017). In concluding their study of 11- to 15-year-olds’ perceptions of
the risks and benefits of using technology, Betts and Spenser (2017, p. 32)
argued “that young people regard technology use as normative” and “they are
aware of the risks of using technology but are overly optimistic with regard the
likelihood of experiencing […] risks.”
6. The combination of important factors, i.e. the huge cost of unhealthy lifestyles to
the NHS, growing concerns over teenagers’ health and unhealthy lifestyles, the
increasing importance of online sources of health information, the critical
importance of social media to wellbeing, and the vulnerability of young teenagers,
highlights the importance of MEMEotive.

 Written evidence submitted by Loughborough University (SMH0167)

4. Scoping Work
7. MEMEotive builds on the success of the EMOTIVE Project (Sykora et al 2013) which was funded by the Defence Science Technology Laboratory following the
London Riots - as a means of detecting and measuring emotions on text-based
social media content. Furthering this work, MEMEotive aims to understand which
Internet memes (defined as “a piece of culture […] which gains influence through
online transmission” (Davison, 2012, p.122)) on social media gain traction and
how they influence/motivate the health and health behaviours of young teenagers.
8. We have undertaken some preliminary research in this area and are seeking
funding for a larger project. The purpose of both this preliminary work and the
broader MEMEotive study is to gain a better understanding of the nature,
transmission and impact of health knowledge and health messages contained in
Internet memes and shared on social media sites among 13-16 year olds.
9. An Internet meme usually consists of an image with text. The same image can be
re-used to convey different messages. Internet memes are generally viewed as
entertaining but they also represent a body of cultural practice that does not
account for the specific needs and rights of teenagers (Livingstone, Carr, & Byrne,
2016). If Internet memes carry political, corporate or other agendas without
priorities tailored to the needs of 13-16-year-olds then they have the potential to
do harm on a large scale (Wartella et al., 2015). We need to know the types of
health information/knowledge that teenagers are exposed to because social
media is an increasingly central aspect of their daily lives and social interactions.
10. Over a two week period we used the Twitter Search API to gather sample tweets
concerning health. Specifically, we collected any tweets mentioning the hashtags:
#meme, #fitnessaddict, and #flexibledieting (47,200 tweets). The latter two
hashtags are fitness and diet related Internet memes, respectively. We also
collected 29,220 tweets mentioning #tea as a relative control.
11. Our initial study revealed concerns about the perceived lack of emotion
associated with the sharing of memes on Twitter. While the broader project seeks
to explore Internet memes across Twitter, Instagram, and Pinterest, this lower
than expected level of expressed emotion when Internet memes were shared on
Twitter is worrying. Memes have the potential to normalise undesirable
behaviours such as trolling, body shaming and bullying, and a lack of emotion
may be indicative of a larger apathy with regards to such practice. These findings,
which are explored in paragraphs 11 to 15, identified the need for a larger scale
project.
12. The number of tweets containing #meme, over a brief time period, clearly
represents a significant amount of user-generated content on the Twitter platform
alone (see table 1 for an overview). To understand the emotional dimension
around these conversations, we employed an established advanced sentiment

 Written evidence submitted by Loughborough University (SMH0167)

analysis tool, called EMOTIVE (Sykora et al. 2013), which uses a semantic
linguistic model to identify explicit emotions in tweets. From the sample tweets
(i.e. #meme, #fitnessaddict and #flexibledieting) EMOTIVE flagged 2,181 (4.62%)
as containing some form of emotions. EMOTIVE also flagged 2674 (9.15%) #tea
tweets as containing emotion.
13. Given the emotional analysis, and drawing on prior research using EMOTIVE, we
are able to make some tentative observations.
a. Considering that Internet memes can often normalise unhealthy behaviours,
figure 1 highlights that a significantly large proportion of tweets across the
three hashtags tended to contain the emotion of happiness. It is worth noting
that nearly twice as much emotion was expressed in tweets containing #tea
than in those containing #meme, and #fitnessaddict and nearly five time more
than those containing #flexibledieting.

Figure 1 – Proportion of emotions across the eight basic emotions detected by
EMOTIVE
b. A previous study (Sykora et al. 2014), 1,570,303 tweets collected from 28
distinct tweet datasets relating to widely varied topics and events, and found
that out of the posted tweets, on average 12% of them contained explicit
emotions (standard deviation, 9%). Table 1 (“Emotionality %” column), shows
that a very low proportion of tweets related to memes are actually expressing
emotions.
Total
Emotional
Emotionality
Twitter Term
Tweets
Tweets
%
#tea
29220
2674
9.15%
#meme
30757
1561
5.10%
#fitnessaddict
9868
505
5.12%
#flexibledieting
6575
115
1.75%
Table 1 – Overview of Tweets collected over a sample time-period
(i.e., 24th May - 12th June),

 Written evidence submitted by Loughborough University (SMH0167)

c. Whilst the sample for this preliminary research equates to about 5% of Sykora
et al.’s (2014) sample, it does uncover a potential ‘hardening’ of social media
users’ emotions towards memes. As shown in appendix 1, the imagery and
text of Internet memes can be contentious and derogatory. Furthermore, in
ridiculing body shape, diet and fitness there is a worry that we are also
normalising obesity, poor diet and sedentary behaviour. If young people are
buying into the same divisive ways of thinking, and are normalising potential
damaging behaviour, then it is vital that we better understand the nature,
transmission and impact of health knowledge and health messages contained
in Internet memes and shared on social media sites among 13-16 year olds.
d. It is worrying that Internet meme content - which, by definition, has come to
represent the Internet-fuelled propagation of items such as videos, jokes,
rumours and websites (Shifman 2014) – produces a predominate sense of
happiness regardless of the underlying tone or image used (see the appendix
for examples of health-related Internet memes). If this is the broader case
then we run the risk of normalising and accepting ridiculing and stereotyping
of “non-normative”, “fat”, “unhealthy”, “irresponsible”, “at fault” individuals
because of cultural ignorance.
e. If, as Dennett (2017, p.206 original emphasis) suggests, memes are “a way of
behaving (roughly) that can be copied” then both information or
misinformation might be “transmitted or saved under the [belief or] mistaken
presumption that it is valuable”. Like counterfeit money, which is saved and
later spent, misinformation regarding health might be used to the potential
detriment of the young person/persons concerned (Casey, In Press).
14. These results are only indicative, and a much more thorough computational,
discourse and semantic analysis of these conversations is needed. Nevertheless,
the above provides some useful and revealing insights into the Internet meme
related social media chatter on Twitter.
15. In addition to more thorough analyses, substantially more context specific
information relating to particular users (i.e. age, ethnicity, gender etc.) posting
Internet meme related content, is needed. This information, coupled with an
appreciation of how individuals react to health-related Internet memes and what
immediate observable impacts these actually have on users, with respect to
particular affected health related behaviours, will allow us to better understand the
impact Internet memes are having on people’s health. Our work in this area will
allow us to delve further into the impact of Internet memes on social media, with
particular reference to 13-16 year olds and their health. We seek to understand
the ways in which young people share/comment on such Internet memes and the
impact this might have on their health and their health choices.

5. Conclusion

 Written evidence submitted by Loughborough University (SMH0167)

16. With the prevalence of social media as a source of health knowledge among
young people, and indications that Internet memes may be playing a part in a
general apathy towards behaviours that ridicule individuals and groups who
display “non-normative", "fat", "unhealthy", "irresponsible", "at fault"
characteristics, the risks that this poses to future generations and our youth are
noteworthy. This kind of social media content has largely gone unnoticed, and its
effects, impact, prevalence and virality are, at best, poorly understood. It is hence
important to investigate this area and seek to impact on social policy initiatives
(such as those that emerge from this inquiry) as the potential impact of Internet
meme appears to be harmful and yet this harm is hidden in images and text. If an
Internet memes is “a piece of culture […] which gains influence through online
transmission” (Davison, 2012, p.122) then we need to be increasingly mindful of
the type of culture we wish young people to see, read and believe. We have it in
our gift to make the world a better place but only, in this case, if we challenge
undesirable behaviours such as trolling, body shaming and bullying that seem
inherent in many Internet memes.

6. References
Betts, L.R. and Spenser, K.A. (2017). “People think it’s a harmless joke”: young
people’s understanding of the impact of technology, digital vulnerability and
cyberbullying in the United Kingdom, Journal of Children and Media, 11:1, 2035, DOI:10.1080/17482798.2016.1233893
Casey, A. (In Press). The Role of Internet Memes in Shaping Young People’s
Health-Related Social Media Interactions. In V.A. Goodyear and K.M. Armour
(Eds.) Young People, Social Media, Health and Wellbeing. London:
Routledge.
Davison, P. (2012). The language of Internet Memes. In M. Mandiberg, ed., The
Social Media Reader. New York: New York University Press., pp. 120–136.
Health and Social Care Information Centre. (2015) Health Survey for England 2014.
Health and Social Care Information Centre. Accessed on 18th July 2018 from
http://webarchive.nationalarchives.gov.uk/20171010184802/http://content.digit
al.nhs.uk/catalogue/PUB19297/HSE2014-Data-Qual-Statement.pdf
Jenkins, H., Ford, S. and Green, J. (2013). Spreadable Media: Creating Value and
Meaning in a Networked Culture. London: New York University Press.
King’s Fund, The (2017). Healthy Behaviours: Future Trends. Accessed on 18th July
2018 from https://www.kingsfund.org.uk/projects/time-think-differently/trendshealthy-behaviours
Livingstone, S., Carr, J. and Byrne, J. (2016). One in Three: Internet Governance
and Children’s Rights. Innocenti Discussion Paper No.2016-01, UNICEF
Office of Research, Florence.
Sykora, M.D., Jackson, T.W., O'Brien, A., and Elayan S. (2013). EMOTIVE
ontology: Extracting fine-grained emotions from terse, informal messages.
International Journal on Computer Science and Information Systems, 8(2), pp.
106-118.
Sykora M., Elayan S., Jackson T. W., O'Brien A. and Lunen A. v., 2014. Twitter
based Analysis of Public, Fine-Grained Emotional Reactions to Significant

 Written evidence submitted by Loughborough University (SMH0167)

Events, European Conference on Social Media - ECSM 2014, Brighton,
United Kingdom
de Vries, D.A., Peter, J., de Graaf, H., and Nikken P. (2016). Adolescents’ social
network site use, peer appearance-related feedback, and body dissatisfaction:
testing a mediation model. Journal of Youth and Adolescence, 45(1), pp. 211224.
Wartella, E., Rideout, V., Zupancic, H., Beaudoin-Ryan, L., and Lauricella, A. (2015).
Teens, Health, and Technology. A National Survey. Evanston, IL: Center on
Media and Human Development. School of Communication. Northwestern
University
World Health Organization (2016). Health Policy for Children and Adolescents No. 7.
Growing up unequal: gender and socioeconomic differences in young
people’s health and well-being. Health Behaviour in School-aged Children
Study: International Report from the 2013/2014 Survey. Geneva: World
Health Organization Regional Office for Europe. Accessed 18th July 2018 from
http://www.euro.who.int/__data/assets/pdf_file/0003/303438/HSBC-No.7Growing-up-unequal-Full-Report.pdf?ua=1
August 2018

 Written evidence submitted by Loughborough University (SMH0167)

Appendix
Sample Meme Images
A Google Images search conducted 25th July 2018.
#meme

#fitnessaddict

#flexibledieting

 Written evidence submitted by Loughborough University (SMH0167)

mug
wn?snnp

?x

 

#cunuimmc'