©
1997 IEEE. Personal use of this material is permitted. However, permission
to reprint/republish this material for advertising or promotional purposes
or for creating new collective works for resale or redistribution to servers
or lists, or to reuse any copyrighted component of this work in other works
must be obtained from the IEEE.
By choosing to view this document, you agree to all provisions of the copyright
laws protecting it.
[ Index | Introduction | Expositive | Demonstrative | Interactive | Practical | Conclusion | References ]
Learning tools based on
the interactive methodology may look similar to the ones based on animation
and local simulation; however, they are instead characterized by a very
important methodological difference.
Expositive and demonstrative learning tools are designed to show
or demonstrate concepts or system behaviors: the learner is in charge
of controlling presentation or simulation parameters only. Interactive
tools, instead, demand directly to the learner to choose, describe or calculate
something. They therefore imply an evaluation of the learner’s answers.
The need for stimulating
learner’s activity has been discussed already. The results of our experience
indicate that, without a continuous self-training
activity, in our field no serious learning is possible. A weakness of
animation and local simulation is the fact that
a student might go through them without paying the necessary level of attention
to assimilate the topics presented. Instead, the necessity of performing
a series of actions in response to requests made by the courseware, typical of
the interactive methodology, demands
a direct learner activity and, consequently, a deeper involvement in the
study. The challenge to accomplish what the tool asks for is generally
working in favor of learner’s motivation [19].
On a programming point
of view, the interaction just described is implemented usually in very
straightforward way, by asking the student to choose, among several options
available, the right one, that is determined by the nature of the problem
under consideration and "known" by the system. Typical example
of one of such problems with a unique solution is the tracing, step by
step, of a timing diagram of an existing logical network, that the courseware
compares with a stored one. Attempts are made to tackle problems with more
than one approach, such as the conception of software algorithms or state
machines diagrams, by pre-defining several solution paths, instead of one.
It must be stressed, though, that approaches based on expert systems techniques
are not taken into consideration.
Most of the interactive
tools described in this chapter have been designed and developed taking
into account the intuition and experience of our lecturers, who have translated
into courseware tools the practical, operational answer to the most important
needs of the learners [20]. The interactive
methodology fulfills several pedagogical needs of the learning environment
that are described in the following paragraphs. It is, in fact, used to
test regularly the proficiency level, to develop tutoring exercises, and
to introduce design training [21].
In our course, short,
rapid multiple-choice and visual tests are one of the implementations of
the interactive methodology. An attempt has been made to structure the
material upon which the test is based. The basic unit is the "question".
The system is composed by standard multiple-choice questionnaires (MCQ)
and interactive questions where the learner is requested to perform a task.
For each action, the system provides a feedback in term of assessment of
the action, help and explanation.
Several questions unified
by a common subject and a quasi-uniform level of difficulty, make a "level".
Access to the next level is granted to the student that has reached a score
above a certain threshold, established by the teacher. At the end of verification,
the system evaluates the candidate with a final score, providing also detailed
information and time spent on each question. Therefore, the auto-evaluation
system, that has been built for all the relevant subjects of the course,
allows students to evaluate their knowledge level and, at the same time,
to consolidate their understanding of digital electronics.
The example shown in Fig.
11 is a classical MCQ whose main purpose is the establishment of a
common language to describe digital systems. The interface is common to
all the evaluation system: the buttons belonging to the bottom bar control
the navigation within the test and the score is indicated in red on the
top right on the screen.
More demanding the test
of Fig. 12, that asks to identify the situation where
the clock frequency is too high for given values of t1 and t2.
The obvious approach of the MCQ has been substituted by an operational
procedure. The learner has to set the values of t1 and t2 with
the proper buttons (center of the screen). The timing diagram changes accordingly
and the condition object of the test has to be identified visually. In
comparison with MCQ, this kind of tests forces the learner to interact
with the timing diagram and to rely more on non-verbal skills. A
working example of the auto-evaluation system described
here is available.
This series of tools is the result of an effort targeted to foster understanding of fundamentals issues that our experience has shown particularly difficult for the beginner to master [22]. Most of these problems are present in the chapter of the course dedicated to the design of digital systems controlled by FSM. The tools developed so far deal with the following issues:
All the tools use a
standard interface and, even if they deal with different issues, they are
characterized not only by the same interactive methodology,
but also by a similar pedagogical sequence. They all start with a statement
of the problem, followed by an animated explanation that, at the beginning
works as "instruction manual" for the tool and then provides,
step by step, the full solution of the exercise. The learner has full control
of the animation and may leave it at any time to start solving alone the
exercise using the interactive part of the tool. In this second part, she/he
does not get any more explanation, but the work is checked step by step
and mistakes are pointed out and counted against the score. Switching between
the two parts of the tool is always possible. We provide in the following
few examples of interactive tutoring exercises, supported by working demonstrations.
The tool of Fig.
13 is dedicated to train the learner in the analysis of an existing
logical network, an important skill for a digital designer. It guides the
learner to understand the behavior of a sequential network composed by
three D-type flip-flops and an EXOR gate. The target of the tool is the
definition of the network timing diagram, given the input signals.
In the animation mode, after an introduction to the
use of the tool, a practical method to perform the analysis is presented
and applied in full. In the solution mode (interactive) the learner practices
on the construction of the timing diagram, using the buttons on top-right
of the screen to define of the three outputs for each clock cycles. When
the learner finds it difficult to proceed in the diagram construction,
she/he may go back, using the button with the open eye, to a complete animated
explanation of each step (working example).
The relations between state
sequence and time evolution is, sometimes, an issue difficult to grasp
by the beginner: courseware tools, taking advantage of animation and
interaction features, may do a better job than traditional textbooks, limited
to static images and textual explanation. The relation between an ASM chart
and its timing is explored using the tool of Fig. 14,
whose purpose is to guide the learner to build the timing diagram (bottom)
corresponding to a given ASM (top). The pedagogical approach is the same
of the previous example: the animation mode explains the behavior of the
simple synchronous FSM, the solution mode allows students’ practice (working
example).
A major problem in digital
design is to train a learner in the conception of a system. This usually implies
the construction of a proper algorithm that in our case can be a state
machine, a software program or a hardware architecture. Such task is difficult
to teach because it requires, in addition to the knowledge of the design
rules, the conception of something that did not exist before. To make a
comparison (a little exaggerated, indeed) it is the same difference as
from knowing the rules of the grammar and writing a meaningful text.
The basic philosophy used
for implementing our interactive algorithm construction is to divide the
algorithm in a ordered set of steps, and to let the learner choose among
several possibilities for each.
These choices are presented
to the student in the form of multiple-choice questionnaires (MCQ): identifying
the right one means to define one algorithm step. When the interactive
tool is developed, the most likely questions that might arise at the different
steps of the exercise are identified. In the case of wrong answers, text
fields and graphics explaining why the answer was incorrect are shown.
A score is generally computed, that is not used for "grading"
but rather to determine whether the student needs more work on the specific
problem. Interactive exercises are generally built upon a unique solution
of a problem. This limitation is acceptable when the purpose of the exercise
is to guide the learner toward a specific solution. Nevertheless, training
for real design implies the removal of this limitation: in our courseware
an unlimited freedom of design is achieved using the general purpose tools
described in the following chapter. It must be understood, tough, that
this implies to renounce the "guide" function provided by this
methodology. In a few cases, like the following example, we’ve been able
to implement interactive exercises, as described above, that allow the
learner to choose among a set of alternative solutions. In other words,
choices made during the interactive exercise determine the construction
of a particular ASM among the ones implemented by the author. To stimulate
learner’s attention (to say it differently, to fight the "video game"
attitude), the tool sometimes accepts wrong solutions for a while and then
warns the learner.
Fig. 15
shows a frame of the interactive tool for FSM construction. The process
starts from design specifications and schematics of the system (not shown
here). Students builds interactively the ASM chart by choosing the right
ASM block from a set of available ones (on the left in the figure). In
this particular case the blocks are dragged with the mouse and dropped
in the workspace in the middle of the screen where, if the choice is right,
they are automatically attached to the chart under construction. A window
appears with explanations and comments. If the choice is wrong the learner
is notified by a window with a text explaining the mistake. Buttons present
on the ASM construction screen allow to recall support information like
the timing diagram or the problem text and schematics. Information, such
as data sheets of digital components used in the system, an introductory
explanation of the problem and other items, are part of the tool (working
example).
The same interactive methodology
is applied to the development of computer programs (in our course mostly
in Assembly language). Instead of ASM blocks, instruction are presented
as multiple choice.
Fig. 11. A multiple-choice question on digital systems at the most elementary
level of difficulty.
Fig. 12. An interactive test evaluates the understanding of the effects
of delays on the operation of a FSM.
Fig. 13. Tutoring exercise assisting the tracing of the timing diagram
of a sequential network.
Fig. 14. Tutoring exercise for training the learner in establishing a correspondence
between state (top) and timing (bottom) of a FSM.
Fig. 15. Interactive construction of a FSM with "drag and drop"
functions.
