An Analysis of Teaching Informatics by Means of Enactive-Haptic

Representations

Lisa G

obel

, Lutz Hellmig and Alke Martens

Institute of Informatics, University Rostock,

Albert-Einstein-Straße 22, 18059 Rostock, Germany

Keywords:

Informatic Concepts, Enactive-Haptic Representation, Teachers, Computer Science and Media Education,

Computer Science Unplugged.

Abstract:

The subject of computer science is gaining more and more important. But how are computer science concepts

taught? The use of enactive-haptic representation can be an option, which is however quite unknown in

computer science education. Partly, this is due to the fact that enactive-haptic is a blurred concept, which is

not easy to grasp in the context of digital technology. In this paper, the term enactive-haptic is analyzed in the

context of computer science. The resulting model and its usage are sketched in this paper.

1 INTRODUCTION

Nowadays, a lot of people call for education in the

ﬁeld of Artiﬁcial Intelligence (AI) in school educa-

tion, accompanied with the claim that kids should

learn up-to-date technology (C. A. Heinze, 2010).

However, we have learned in different investiga-

tions, that ﬁrst of all, teachers are not enough educated

to teach AI, and second, AI in itself is not a technol-

ogy but an application of technology. Our opinion is

that schools should rather educate kids in the ﬁeld of

basic science insights and not hot topics, which occur

and vanish over the course of time (just remember: AI

has been a big thing in the 1970s).

Instead of marveling about the ”next big things”,

students should understand timeless, fundamental

ideas (Schwill, 1993) of computer science (or in-

formatics, which in the German version is the most

used term in Germany). To allow education of the

fundamentals of computer science, an agreement has

to be found, what these fundamentals are. In Ger-

many, the politicians and Universities who educate

computer science school teachers have a while ago

agreed to a certain form of curriculum, which we

call the Rahmenplan. Herein, it is ﬁxed that shall

be taught when and how. As modern technology de-

velops over time, the University of Rostock and the

country Mecklenburg-Western Pomerania have just

https://orcid.org/0000-0003-3949-1022

https://orcid.org/0000-0002-9411-920X

recently agreed to make a modern form of this cur-

riculum (Rahmenplan), which is putting the ﬁnger on

the set of fundamental concepts, which very likely

will be the main corpus of computer science in the

next twenty years. Thus, we think it is currently more

important in our perspective, that school kids under-

stand what an algorithm is, instead of learning about

AI (which in the core is also algorithms). Moreover,

after so many years of computer science school ed-

ucation (we had this for over 30 years), we are sure

that is useful to unlock from current tools. Thus,

one insight resulting from this approach is that com-

puter science education works best when the underly-

ing concepts can potentially also be understood with-

out any tools! This is called the unplugged approach

(Gallenbacher, 2017).

However, even the unplugged approach does not

come without didactic or instructional design. Thus,

we looked closer at the different ways to offer ma-

terial to school kids on the best interactive level.

Here we ﬁnd the traditional educational idea of dis-

tinguishing between enactive-haptic, iconic and sym-

bolic (Bruner, 1970).

The use of enactive representations could be an

appropriate way to get a deep understanding of ab-

stract ideas. The use of enactivities in computer sci-

ence has already been taken up by Bell (T. Bell, 2005)

– for example Treasure Island or The Orange Game

(T. Bell, 2005) – and Gallenbacher (Gallenbacher,

2017) – for example Binary Magic. Gallenbacher also

hosted an exhibition on enactivities in computer sci-

Göbel, L., Hellmig, L. and Martens, A.

An Analysis of Teaching Informatics by Means of Enactive-Haptic Representations.

DOI: 10.5220/0011765200003470

In Proceedings of the 15th International Conference on Computer Supported Education (CSEDU 2023) - Volume 2, pages 207-212

ISBN: 978-989-758-641-5; ISSN: 2184-5026

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

207

ence. (Hugh, 2010) At the University of Rostock, a

pilot project was developed in which student teach-

ers ”present computer science in an exciting way”

(L. G

obel, 2019b) by using enactive-haptic represen-

tations.

However, investigating how teachers work in

school and prepare their material, we were surprised

to learn that the enactive-haptic approach is seldom

used. The same was also observed some years ago by

other researchers. In 2011 Kalbitz et al. published a

paper about the use of the (Enactive-Iconic-Symbolic)

EIS principle in computer science lessons by Berlin

teachers. The data from 40 surveys showed a lack of

using enactive representations, although the EIS prin-

ciple was known by the teachers. Enactive methods

were used less by the respondents than iconic meth-

ods. One reason was, that the teachers hardly knew of

any alternatives (M. Kalbitz, 2011).

In the beginning, a literature study was conducted

to determine the current state of research on enactive-

haptic representations. After that, we came up with

different categories or model of enactivity, which are

described in this paper. We tried to make an order

between the categories and the existing approaches,

which was successful and will be described in the fol-

lowing.

2 THE ROOTS OF

ENACTIVE-HAPTIC

REPRESENTATION

First of all, it has to be clariﬁed what enactive-haptic

representation really is. As a second step, a survey

will be conducted in which the focus will be on the

use of enactive-haptic representations.

The concept of enactive-haptic representations in

the classroom is based on the psychology of learn-

ing. Three main steps that over time led to the idea of

enactive-haptic learning, are brieﬂy outlined.

The ﬁrst well-known approach can be traced back

to Piaget described an approach to the cognitive de-

velopment of children through the theory of genetic

epistemology (genetic in this context means a kind

of development, in this case, the cognitive develop-

ment, but also the general epistemic development in

human beings). Depending on the age of the child,

four phases are distinguished – sensorimotor phase,

preoperational phase, concrete-operational phase and

formal-operational phase. ”Piaget was convinced that

children ”construct” their schemata through their in-

teraction with the environment”, said Mietzel (Miet-

zel, 1998). Thus, Piaget was one of the ﬁrst who fo-

cuses on interaction with material as a means of learn-

ing, which directly leads to the creation of cognitive

relations (admittedly, the concept of cognition stems

from a later period, as Piaget’s ﬁrst works were all

related to behaviorism). As the second in the row,

Aebli paid more attention to the process of educa-

tion and teaching and abstracts from the age-speciﬁc

model. He described three main stages: the concrete,

the pictorial and the symbolic stage. In the concrete

stage, work is done with concrete objects and mate-

rials. In the next stage – the pictorial stage – objects

represented pictorially are operated with. In the sym-

bolic stage, objects and operations are represented by

signs. The stage transitions are realized by reﬂecting

on one’s own activity, verbalizing the action, or by

practice (Aebli, 1985).

This step-wise approach by Aebli was taken to

a new level by Bruner, here the third one in our

short history of ”enaction”. With his enactive-iconic-

symbolic-principle (EIS), Bruner describes the differ-

ent ways of representing knowledge and skills. A

distinction is made between the enactive, iconic and

symbolic levels of representation. The modes of rep-

resentation refer to each other reciprocally and exist

on the same level (Bruner, 1970).

And now, the relation becomes clear: enactive

means in this sketched tradition that learning is best

taking place by using the human senses to interact

with the learning material. Enactive-haptic thus fo-

cuses on the usage of the sense haptic in the context

of learning, which mainly means: touching and inter-

acting with the material.

Coming to this point, we had to ask ourselves, how

this can be realized in Computer Science education?

This question has been answered by several other re-

searchers before, mainly in realizing approaches like

the unplugged model or other comparable ”hands-on”

material. But still, the concept is somewhat hard to

mediate to teachers-to-be. What confused them can

be mainly reduced to aspects like what is meant by

haptic in computer science is imitating the teacher’s

enactive, must the kid ﬁnd it out all by themselves,

how much information must or can be embedded in

the material, what can I do if a kid is stuck?

So, the process of clarifying the concepts of en-

activity and representation requires a more detailed

analysis. What is required is a collection of ways

the school kids (the students) interact with the ma-

terial, which also represents the way a learning gain

is made, and which its core represents the advantage

of the method in relation to other methods. The inter-

action level can contain attributes like imitate, try out,

explore, listen, watch, try out and listen and the like.

Additionally, a spacial level exists: here the distinc-

tion can be made by answering the question whether

the room plays a role or not, and if it plays a role, then

CSEDU 2023 - 15th International Conference on Computer Supported Education

208

how can this be realized? We located two major levels

in the investigated enactive-haptic examples, which

led us to the distinction between the immediate level

and the mediate level. The mediate level can itself be

further separated into two additional levels, which are

the ﬁrst-order hybrid representation and the second-

order hybrid representation. These will be explicated

in the following.

The immediate level consists of the interaction

level, which are touching, interaction with things, etc.

Learning takes place by doing and the spacial level is

given by the radius a student has for the interaction

types. Entirety, minimally invasive and interpersonal

representations are part of the immediate level.

The mediate level consists mainly of something

we would call virtual. With this, we mean everything

where the student has to internally visualize things

without direct haptic feedback by an artefact. For this

purpose, the student uses his or her auditive and visual

sensory systems but has nothing to touch. The medi-

ate level describes enactivities that are less direct. Re-

garding spatial representation, the objects referred to

are not part of the student’s environment. The com-

puter can be a means of visualization, thus, the repre-

sentation comes in form of a medium. Hybrid repre-

sentations of ﬁrst-order and second-order are part of

the mediate level. Speciﬁcally, the students get visu-

ally in touch with the activity. Every activity from the

immediate level can convert into the mediate level.

Figure 1 shows the different levels with their

meaning.

Figure 1: Representation of enactivity.

3 USING ENACTIVE-HAPTIC

REPRESENTATION IN

COMPUTER SCIENCE

EDUCATION

This research examines the current state of the art of

teaching computer science (or informatics) concepts

by means of enactive-haptic representations.

3.1 Entirety Representation

View two examples ”Treasure Island” by Computer

Science unplugged (T. Bell, 2005) or ”Von-Neumann-

Principle experience” by G

obel, Hellmig (L. G

obel,

2019a).

Treasure Island is a ”game” to teach the concept

of a ﬁnite state automaton. The classroom is con-

structed in a way that it represents different islands

in an ocean. Pirate ships are sailing from island to is-

land. The pirates take travelers – the students. That

way the students want to sail to Treasure Island. From

each island, only two different pirate ships are sailing.

The students get a map on which they draw the route.

(T. Bell, 2005)

Figure 2: Part of the result map (T. Bell, 2005).

The result of the map shows a ﬁnite state automa-

ton with two state transitions. After that activity, the

teacher and students discuss the result and formal-

ize the deﬁnition of a ﬁnite state automaton (T. Bell,

2005).

The experience of the von-Neumann-Principle is

a ”theatrical performance”. The classroom symbol-

izes von Neumann’s architecture. The students are

elements of this architecture. They run through all

stations. In that way, the students run a program of

von Neumann architecture. They see that all tasks are

processed one by one. So for example they can di-

rectly experience the von Neumann bottleneck. The

teacher moderates this activity (L. G

obel, 2019a).

Figure 3: Classroom of the Von-Neumann-Principal experi-

ence (L. G

obel, 2019a).

An Analysis of Teaching Informatics by Means of Enactive-Haptic Representations

209

After that ”theatrical performance”, the class summa-

rize the concept of the von Neumann principle as a

fundamental idea of computer science.

Both examples have a similarity. This representa-

tion describes the classroom or environment of the

students as a part of the interaction dimension. The

students are also a part of the interaction dimension,

as they play the role of an entity of the ”simulation”.

Mostly the teacher is a moderator or observer. Af-

ter the ”theatrical performance”, the teacher (together

with the students) evaluate the results and explicated

the underlying computer science concept. That enac-

tivity is called entirety representation in our analysis.

It is taught in an immediate way. Things are directly

(haptic) explored and bodily experienced.

3.2 Minimally Invasive Representation

The next examples are the magic trick of binary

system by Gallenbacher (Gallenbacher, 2017) and

text compression by Computer Science unplugged

(T. Bell, 2005).

The magic trick is a paper model sheet of 36 cards

with an ”X” and on the other side, an ”O”. A stu-

dent orders 25 cards of the sheets in a random square.

The last 11 cards are placed in a bigger square by the

teacher or wizard. The teacher or wizard looks away

while the students pull one card around. After that,

the teacher or wizard shows the inverted card. The

audience has to ﬁnd out why the wizard knows that.

The students have to develop their own theory about

the trick. They have to check how many ”X” and ”O”

are in a row and column. In this way, the students

learn error detection and how many errors can be de-

tected. (Gallenbacher, 2017)

Figure 4: Error detection (Gallenbacher, 2017).

Error detection is included in the checksum algo-

rithm. For example, credit cards have a certain num-

ber. This number includes a checksum algorithm. Af-

ter ﬁnishing the test, these aspects are highlighted by

the teacher who resumed the computer science expla-

nation.

The second example is based on the limit of infor-

mation. An efﬁcient way to save or send information

is compression. The activity starts with a poem. As a

ﬁrst step, the words can be compressed by putting ar-

rows between identical patterns. In the next step, the

students complete a text by using the pattern. So the

text is compress(T. Bell, 2005).

Figure 5: Text compression (T. Bell, 2005).

The compression of data is an important fact of data

transfer, for example, the pictures in social media are

compressed. Same as above, this is summed up after

the experiment by the teacher and brought together to

mediate the computer science concept.

The similarity of both examples is the way the ma-

terial is used. The students edit the material at their

desks. They can interact with haptic material individ-

ually or in a group. The teacher is a moderator or

observer. This form of representation is called mini-

mally invasive representation. It is teaching in an im-

mediate way. The main difference to the ﬁrst example

is that the students are not part of a ”simulation”, and

the haptic information has no additional bodily expe-

rience level.

3.3 Interpersonal Representation

Examples of interpersonal representation are the net-

work by Helfrich (Helfrich, 2017) and the Orange

Game by Computer Science unplugged (T. Bell,

2005).

The network by Helfrich uses pegs and ﬂashcards.

The students act as different routers which want to

communicate with each other. So they write a ﬂash-

card with sender, addressee and a message. The ﬂash-

card is ﬁxed with a peg on a thread. The network

grows after every iteration. So the students learn the

basic idea of communication between routers. (Hel-

frich, 2017)

This enactivity version shows different topologies

of networks and how they route information. Real

networks are built according to a certain topology, de-

pending on their task. This information is provided

again by the teacher after summing up the results.

Usually, information about what a router is is given

CSEDU 2023 - 15th International Conference on Computer Supported Education

210

Figure 6: Network design with peg and ﬂashcard (Helfrich,

2017).

in a lesson before the enactive work.

The Orange Game mediates routing and deadlock

in networks. A group of students sits in a circle. Each

student has a number. There are two oranges (indeed

the fruits are meant. Balls might do the same.) with

the same number for each student. The oranges ran-

domly distribute to the hands of the students. One

student has only one orange. The students pass the

oranges to their immediate neighbours till every stu-

dent has their member oranges. (T. Bell, 2005)

The students learn that they have to work together

and not be ”greedy”. (T. Bell, 2005) The fundamental

Figure 7: The Orange Game (T. Bell, 2005).

idea of that game is the deadlock. The deadlock is

part of processes.

These examples focus on the haptic exchange of

things between students. The environment of the stu-

dents does not matter, thus it is a combination of en-

tirety representation and minimally invasive represen-

tation. The students are a part of the activity. But

the activity does not ﬁll the complete room. The stu-

dents can get allotted materials. This representation is

called interpersonal representation.

3.4 Hybrid Representation First-Order

The next example is the treasure hunt by RWTH

Aachen University. A website includes a simulation

to ﬁnd a treasure. The necessary material is hidden

in different towns which are positions in the class-

room. The students solve encodings. In that way,

they learn how different coding systems work. The

simulation at the computer supports the treasure hunt.

(N. Bergner, 2020) The treasure hunt shows different

Figure 8: Treasure hunt (N. Bergner, 2020).

symmetric encodings. Encoding is an important pro-

cedure to keep messages secret. The students learn

how secure that procedure is.

That enactivity has a different level than the for-

mer examples. The computer is included in that ac-

tivity. So, the level is switched. It is on a mediate

level. Because of the shift of the haptic, it is a hy-

brid representation of the ﬁrst-order. The ﬁrst-order

hybrid representation is a combination of the use of a

computer and haptic material.

3.5 Hybrid Representation

Second-Order

The last example is the movie ”Technology on the In-

ternet” by dandelion. The clip shows a haptic option

of run-length encoding. There are little men who call

the numbers ”zero” or ”one” dependent on the encod-

ing. So the clip shows how a picture will send from a

scanner through the internet. (ref, 2022)

The idea of that ﬁlm shows an option of a com-

press to transfer data.

The ﬁlm shows a haptic representation. This hap-

tic is produced in the mind of the audience. They

can imitate that encoding. That representation is at

a mediate level. It is a combination between a video

recording and an immediate level activity. So it is

called hybrid representation second-order.

4 OUTLOOK

The term enactive-haptic representation has been the

starting point of the research sketched in this paper.

We have started with the insight, into why enactive-

haptic realizations are so seldomly used in classroom

settings. We came to the point that the overall con-

An Analysis of Teaching Informatics by Means of Enactive-Haptic Representations

211

cept might be too coarse-grained when it comes to

computer science, as the main parts of the subject are

abstract and not easy to grasp in a hands-on scenario.

After the ﬁrst steps towards the categorizations,

we resulted in different instructional (or didactical)

models of education based on the enactive-haptic per-

spective. In each of the above sketched different ap-

proaches, it becomes quite clear, where interaction

takes place, which role the spacial dimension has,

and which forms of enactivity can be realized. The

yet existing examples for educational scenarios all ﬁt

into the schema, as shown above. Starting from this

point, all the necessary ingredients for developing dif-

ferent types of enactive lectures are available and thus

might potentially be used for instructional design in

computer science. However, what is still missing is

the reality check. Thus, we are currently developing

a questionnaire and potentially interviews with com-

puter science teachers at schools. In contrast to Kalb-

itz et al. (M. Kalbitz, 2011) we will operate more dif-

ferentiated by categorizing the enactivity and a higher

number of cases.

When investigating the form of interaction the fo-

cus of the questionnaire is on the following ques-

tions: ”What is the state now?”, ”If enactive-haptic

representations are used, why and how?” The term

”enactive-haptic” will probably be unknown, so the

query must be made via other questions or methods.

The scope of the teachers’ survey should include a

spectrum of teachers as broad as possible, which al-

lows a cross-sectional and longitudinal approach.

In the following, the methodological procedure

will be worked out. Criteria for determining a se-

lection of subjects must be formulated. First, there

should be a preliminary study, from which a guided

interview can be derived. Second, a survey is to be

conducted in Mecklenburg-Western Pomerania, since

this was the ﬁrst federal state to introduce the compul-

sory subject of computer science and media education

(R. Schwarz, 2022). It is possible to extend the survey

to other federal states with the help of a questionnaire.

Subsequently, an evaluation can be carried out, which

should lead to the formation of hypotheses. An out-

look for further research can be worked out. Further-

more, the results should show where further training

is needed.

Another next step will be to detect how to educate

teachers and teachers-to-be (i.e. students) in using

enactive-haptic representations and realizations. One

option could be targeted training and vocational train-

ing for teachers, where we have a broad experience in

our lab.

REFERENCES

(2022). Technik im netz. ZDF: L

owenzahn.

Aebli, H. (1985). Zw

olf Grundformen des Lehrens. Klett-

Cotta, Stuttgart, 2nd edition.

Bruner, J. S. (1970). Der Prozess der Erziehung. Berlin

Verlag, Berlin.

C. A. Heinze, J. Haase, H. H. (2010). An action research

report from a multi-year approach to teaching artiﬁcial

intelligence at the k-6 level. In First AAAI Symposium

on Educational Advances in Artiﬁcial Intelligence.

Gallenbacher, J. (2017). Abenteuer Informatik. IT zum An-

fassen f

ur alle von 9 bis 99 – vom Navi bis Social Me-

dia. Springer, Heidelberg, 4th edition.

Helfrich (2017). Rollenspiel netzwerk.

Hugh, A. (2010). Informatik hautnah erleben. Arbeits-

berichte des Fachbereichs Informatik, Nr. 11.

L. G

obel, L. H. (2019a). Die von-neumann-prinzipien

erleben. ein enaktiver zugang zu einem abstrakten

thema. LOG IN, 39(191/1921):49 – 59.

L. G

obel, L. H. (2019b). Informatik spannend pr

asentieren

- ein einstieg in den studiengang lehramt informatik.

In Informatik f

ur alle, pages 191 – 200. Gesellschaft

ur Informatik e. V.

M. Kalbitz, H. Voss, C. S. (2011). Informatik begreifen - zur

nutzung von veranschaulichungen im informatikun-

terricht. In Informatik in Bildung und Beruf, pages

137 – 146. Gesellschaft f

ur Informatik e. V.

Mietzel, G. (1998). P

adagogische Psychologie des Ler-

nens und Lehrens. Hogrefe, G

ottingen, Bern, Toronto,

Seattle.

N. Bergner, K. Kamphausen, D. K. (2020). Schatzsuche.

InfoSphere - Sch

ulerlabor Informatik RWTH Aachen

University.

R. Schwarz, L. Hellmig, S. F. (2022). Informatik-monitor.

Gesellschaft f

ur Informatik e. V.

Schwill, A. (1993). Fundamentale ideen der informatik. In

Zentralblatt f

ur Didaktik der Mathematik.

T. Bell, H. Ian, M. F. (2005). Cs unplugged. In An en-

richment and extension programme for primary-aged

students.

CSEDU 2023 - 15th International Conference on Computer Supported Education

212