Technical Translation: Usability Strategies for Translating Technical Documentation

Declarative Memory 
Declarative memory consists of a number of different types of memory: 
episodic, associative, lexical, image memory and semantic. 
Episodic memory
is our memory of events and facts relating to them 
(Dix 1998:31; Faulkner 1998:37). This memory allows us to reconstruct 
certain events and remember facts about them. Coe (1996:75) also men-
tions a specific type of episodic memory which is like a high-resolution 
snapshot of a particular event that was particularly surprising, emotional or 
otherwise significant. This is known as 
flashbulb memory
and is attributed 
to momentous occasions, be they good or bad. 
Associative memory
is the way we remember information using tags 
Lexical memory 
is what we use to remember the graphical and phono-
logical features of words. This refers strictly to the physical properties of 
words – the combination of black and white lines on paper, for example. 
The m
meaning of these words, however, is stored in 
semantic memory
. 
The term 
image memory 
can be quite misleading because it does not 
only refer to physical pictures we have seen and stored in memory but also 
to mental images which we construct on the basis of events, pictures, situa-
tions, words etc. For instance, we can picture in our minds a childhood 
birthday party or a beautiful place we may have seen on holiday. We can 
also manufacture mental images without ever having seen the physical ob-
ject or place in question. This is part of what we refer to as 
imagination
and 
it is a product of our image memory. For instance, we can picture ourselves 
sipping frozen margaritas on the deck of a yacht in the Caribbean even 
118 
with which we label schemes of knowledge (see also page 35).

---

Memory
though we may have never been on a boat in our lives, nor visited the
Caribbean. 
This type of memory is more durable and reliable than any other type of 
memory (Coe 1996:77). When we store information, either in the form of 
an image or accompanied by an image, we can recall it more readily than 
information that does not have image associations. For instance, it is easier 
to remember where the dipstick is on a car if we associate a visual image of 
its location rather than a verbal description alone, e.g. a long piece of metal 
protruding from the oil sump in a car’s engine. 
Semantic memory
is our memory for facts, meanings, concepts, vocabu-
lary etc. Semantic memory is our knowledge about the world – a structured 
record of facts, knowledge, concepts and skills that we have acquired. Se-
mantic memory is structured to allow orderly and reliable access to the in-
formation (Dix 1998:31). One model used to explain how this information 
is structured is that of the network. Classes are used to relate items together 
and each item may inherit attributes from superordinate classes (Dix 
ibid.
). 
The classes are all linked together in a network. Semantic networks, how-
ever, do not allow us to model the representation of complex objects or 
events. They merely allow us to model relationships between single items 
in memory (
ibid.
). 
Another structure proposed to counteract this problem is the notion of 
frames 
and 
scripts
. This model organises information into data structures. 
These structures contain 
slots
which contain default, fixed or variable at-
tributes. Scripts are an attempt to model the representation of stereotypical 
knowledge about given situations (Dix 1998:33). This representation allows 
us to interpret partial descriptions fully. The frames and scripts are then 
linked together in networks to present hierarchically structured knowledge. 
Another type of knowledge representation is the representation of pro-
cedural knowledge. This is our knowledge about how to do something. 
Procedural knowledge is generally represented in a production system con-
sisting of “
if-then
” condition-action rules (Dix 1998:34). With this model, 
information that is received in STM is matched against one or more of 
these rules and the associated 
action
is determined by the 
then