Cognitive Relativism in Management Accounting

INTRODUCTION
What happens when people make decisions about an accounting phenomenon, amid the
pressures, constraints, dangers, and opportunities of todays business environment? 1
This chapter1presents a model that focuses on the cognitive process employed by a
decision maker attempting to use his/her judgment to make a decision about an
accounting phenomenon. Basically, both judgment and decision are the products of a
set of social cognitive operations that include the observation of information on the
accounting phenomenon and the formation of a schema to represent the accounting
phenomenon that is stored in memory and later retrieved when needed to allow the
formation of a judgment and a decision. Before we present the model, an elaboration on
the notion and use of schemata in cognitive psychology and accounting is necessary.
SCHEMATA IN COGNITIVE PSYCHOLOGY The Notion of Schema-Guided
Processes
The schema theory as developed by F.C. Bartlett2 served as the stimulus for all
schema theories. As defined by Bartlett, a schema is an active organization of past
reactions, or past experiences, which must always be supposed to be operating in any
well-adapted organic response.3 Schemata are complex unconscious knowledge, as
masses of organized past experiences.4 They are generic cognitive representations, in
the sense that they constitute a process that can deal with an indefinitely large number
of new instances.
Modern views of schemata refer generally to cognitive structures that represent
organized knowledge about a given concept or a given stimulus and that serve as
mechanisms for the interaction of old knowledge and new knowledge in perception,
language, thought, and memory.
Schemata are generally regarded as fundamental elements upon which all information
processing depends. They constitute a theory about knowledge: how knowledge is
represented, and how that representation facilitates the use of knowledge in numerous
ways. As stated by D.E. Rumelhart, Schemata are employed in the process of

interpreting sensory data, in retrieving information from memory, in organizing actions,

in the determining of goals and sub-goals, in the allocation of resources and generally
Cognitive Relativism

in guiding the flow of processing in the system. 5 In fact, useful analyses of schemata
suggested by Rumelhart include plays, theories, procedures, and parsers.6 Properties of
schemas include the following:
1. A schema represents a prototypical abstraction of the complex concept it represents.
2. A schema is induced from past experiences with numerous exemplars of the
complex concept it represents.
3. A schema can guide the organization of incoming information into clusters of
knowledge that are instantiations of the schema itself.
4. When one of the constituent concepts of a schema is missing in the input, its
features can be inferred from default values in the schema.7
Schemata versus Categories
Jean Mandler8 made an unusual distinction between two types of represen- tations
categories and schemata. Categories are denoted by verbal or nonverbal symbols (i.e.,
names) and are represented by a set of features that serve as the basis for inferring
membership in it. Schemata, on the other hand, are cognitive representations whose
features, like those of categories, are organized according to specific a priori spatial,
temporal, or logical criteria. Categories and schemata function differently. As Robert
S. Wyer and S.E. Gordon note:
Information about a set of attributes processed by the members of a particular category
may not spontaneously activate this category unless either (a) the attributes are very
strongly and uniquely associated with it, (b) one has a specific objective that leads the
object being described to be classified, or (c) a category and its characteristic features
are already activated at the time the information is received ... In contrast, information
that describes the characteristic features of a schema may become more inclined to activate the schema spontaneously.9
Schema Growth and Change
In considering schema growth and change the evidence favors a perseverance effect
whereby generic schemata are resistant to change even in the face of contrary

evidence.10-11 In fact, people may even interpret exceptions as proving a given schema,
unless they are asked to counterargue it, to explain why their favorite theory might be
wrong.12

Cognitive Relativism

Schemata are developed from experience with instances of the category in question
and become more complex, more abstract, and more organized with experience. With
increasing experience a schema becomes more mature and more complex. Hence, the
schemata of experts contain more informational elements than those of novices, are
more organized, contain more links, and may have a more complex hierarchy.13-15
Sources of Activation for Schemata
D. G. Bobrow and D. A. Norman16 distinguish between two basic sources of
activation for schemata: conceptually driven and data-driven processing. In conceptually driven processing, an activated schema in turn activates a subschema with the
expectation that this will account for some portion of the input data. In data-driven
processing, the activated subschema causes the activation of the various schemata of
which it is a component. Data-driven processing goes from the part to the whole. In
another source of activation, known as schema-directed processing, the activation is
assumed to go in both directions. It proceeds as follows:
Some events occur at the sensory system. The occurrence of this event automatically
activates certain low level schemata (such schemata might be called feature
detectors). The low level schemata, in turn, activates (in a data driven fashion) certain
of the higher level schemata (the most probable ones) of which they are constituents.
These higher level schemata then initiate conceptually driven processing by
activating the subschemata not already activated in an attempt to evaluate their
goodness of fit.17
Encoding of Information in a Schema
For W. F. Brewer and G. V. Nalsamura, the interaction of old knowledge with new
knowledge involves two processes: one refers to the modification of the generic
knowledge in the relevant schema, while the other refers to the construction of a
specified instantiated memory representation, where the instantiated schema is the

cognitive structure that results from the interaction of the old information and the new
information from the episodic unit.18
Cognitive Relativism

The encoding of information is in fact subject to at least two interpretations. First, the
interpretation proposed by R.S. Woodswork and H. Schlosberg19 postulates that once a
schema is activated by incoming episodic information, features that are inconsistent
with the implications of this schema are appended to the representation of information
as corrections. A second conceptualization, proposed by A. C. Graesser, S. E.
Gordon, and J. D. Sawyer,20 known as the script-pointer-plus-tag formulation,
postulates that when people receive information that is interpretable in terms of a
prototypic event schema (script), they do not retain the information itself but retain a
pointer to the general script, along with an indication of the values of the information
that instantiate the script variables. If features of the information do not match attributes
of the generic script, and thus cannot be reconstructed, they are appended to the representations as tags. Basically, new information is represented by a series of
pointers to prototypic event schemata that can be used to understand or describe the
event, accompanied when necessary by tags denoting objects or events that cannot be
derived from the event schemata alone.
Social Schema Research
Social schema research investigates self-schemata, person schemata, script or event
schemata, and person-in-situation schemata.21
The self-schema contains cognitive generalizations about the self that are derived
from past experiences. People are generally self-schematic on dimensions that are of
importance to them, on which they perceive themselves as extreme, and on which they
perceive the opposite to be untrue.22 They are schematic on those dimensions perceived
to be of lesser importance to them.
Research on perception shows that people who are schematic on a particular
dimension recognize and filter rapidly incoming information about the dimension,
notice the dimension in other people, and think harder about kinds of schema-relevant
information.23
Research on memory shows that self-schematic people remember schemarelevant
information, are difficult to change, have more accessible knowledge about others

because of the sheer familiarity of self-knowledge, and are more affect-laden in

knowledge about others, especially unfamiliar individuals.24-25
Cognitive Relativism

Research on inference shows that people make rapid predictions about their own
behavior that are consistent with their self-schemata. 26 Under certain circumstances
these predictions take longer than for aschematics,27 especially if the judgment is novel.
The person-schema contains cognitive generalizations about traits and behavior
information common to certain groups or types of people.
Research on perception shows that categories for people, like categories for objects,
are organized hierarchically.28 Research on memory shows that schemata for peoples
traits and goals typically help the perceiver to remember schema-consistent information
in more detail than would be possible without the schemata. Research on inference
shows that person schemata affect subsequent inferences.
The evidence on person schemata is summarized as follows:
Person schemata include protypical representations of traits such as extroversion and
introversion, as well as notions of what behavior is consistent with a given goal. Person
schemata of all sorts shape the processes of perception, memory, and inference to conform to our general assumptions about other people. The effects of schemata on perception, memory and inference are not necessarily well suited to accuracy in identifying
individual instances. Schemata are used by the mind to manage such processes economically, if not accurately.29
The script or event schema contains cognitive generalizations that describe the
appropriate sequence of events in a given situation. 30 Research on script or event
schemata is summarized as follows:
Script or event schemata describe sequences of activity from everyday life. They
contain props, roles and sequence rules. Scripts also may be subdivided into segments
(scenes). Like other schemata, scripts guide the perception of ambiguous information
and often shape memory toward schema consistent information. Inferences can be seen
as filling in gaps where information was missing, and gap filling appears to be
exaggerated by repeated encounters with the script. Most of the functions of scripts echo
those of other schemata, in their focus on relevantand usually on consistent
information in perception, memory and inference.31

The person-in-situation or role schema contains cognitive generalizations about

people in situations or scripts for behavior in situations. Role schemata not only help
Cognitive Relativism

perception, memory, and inference but may be a way to account for stereotyping.
Research on perception shows that categorization instantiates the stereotypic content
of the schema whether or not the person fits the category and in the process minimizes
the amount of variability and complexity that may exist in the category. 32-33 In addition
to minimizing variability and complexity, a schema slants perception of the content of
what a person does.
Research on memory shows that the role schema shapes memory in a schemaconsistent fashion. In addition, the categorical information seems to override the details
of the specific instance.34 Schema-discrepant information is, however, likely to receive
added attention at input, if task conditions allow. Attentional processes can facilitate
remembering inconsistent information.35
SCHEMA-BASED RESEARCH IN ACCOUNTING Cognitive Research in
Accounting and Auditing
The cognitive revolution in social psychology has created strong interest in the
knowledge structure in memory in general and how people learn in particular. This
research paradigm also affects accounting and auditing. Given that the difference
between declarative knowledge and procedural knowledge is equivalent to the
difference between content knowledge and the use of that knowledge or between
knowing what and knowing how, W.S. Waller and W.L. Felix used the concepts to
propose a model of how an ordinary person learns from experience. 36 Basically, its
thesis is that learning from experience involves the formation and development of
generalized cognitive structures that organize experience-based declarative and
procedural knowledge in long-term memory. Declarative knowledge is organized by
categories, which depend on similarity of class membership relations, and schemata,
which depend on spatial and/or temporal relations. Procedural knowledge is organized
into production systems, i.e., hierarchies of condition-action pairs.37
What the model implies is that schemata are developed through a gradual process of
abstracting domain-specific knowledge on the basis of experience. The difference
between the expert and the novices knowledge structure is therefore the result of

difference in experience. What is apparent from the research on novices and experts is
that longer chunks of information are taken and stored by experts than novices at any
Cognitive Relativism

point in time and for a particular task; 38, 39 pieces of information are better clustered into
meaningful categories within a single chunk by experts; 40 and the recall of experts is
based on conceptual representations for information while the novices is based on
functional relationships.41-44
The findings in accounting so far parallel those in psychology. More specifically, R.
Weber found that expert auditors clustered internal control cues according to their
control categories significantly more than novices did.45 D. M. Frederick and R. Libby
found that expert auditors clustered financial statement errors by transaction cycles.46
Propositions about Knowledge Structures in Accounting
The notion of schemata (knowledge structures or templates) was used by Michael
Gibbins to make general propositions, corollaries, and hypotheses about the
psychological operations of professional judgment in the natural everyday settings
experienced by public accountants.47 Professional judgment in public accounting was
described as a five-component process:

schemata or knowledge structures accumulated through learning or experience

a triggering event or stimulus

a judgment environment

a judgment process

a decision/action
The list of propositions, corollaries, and hypotheses is proposed. While it awaits
empirical validation, the list constitutes one general descriptive theory of professional
judgment in public accounting, where auditor judgment is viewed as a responsive,
continuous, unconscious, instrumental process of sequentially matching cues to
knowledge structures to generate preferences and responses based on experience. 48
Preliminary findings on these propositions are provided by Gibbins, 49 Gibbins and
Emby,50 and Emby and Gibbins.51
A COGNITIVE VIEW OF THE JUDGEMENT/DECISION
PROCESS IN MANAGEMENT ACCOUNTING

In this section a model of the judgment/decision process in accounting is proposed as

an exercise in social perception and cognition, requiring both formal and implicit
Cognitive Relativism

judgment.52 The primary input to this process is an accounting problem or phenomenon

that needs to be solved and requires a judgment preceding either a preference or a
decision. The model consists of the following steps:
1. Observation of the accounting phenomenon by the decision maker
2. Schema formation or building of the accounting phenomenon
3. Schema organization or storage
4. Attention and recognition process triggered by a stimulus
5. Retrieval of stored information needed for the judgment decision
6. Reconsideration and integration of retrieved information with new information
7. Judgment process
8. Decision/action response
Observation of the Accounting Phenomenon by the Decision Maker
The decision maker is assumed to have the opportunity to observe the accounting
phenomenon. To understand the accounting phenomenon, the decision maker may be
given some information that is deemed diagnostic. If this information is not provided,
the decision maker may seek the information and test available information judged
most relevant to the phenomenon. Following H.
H. Kellys approach to causal attribution,53 the search behavior may concentrate on
these types of available information:
I. Consensus information: how this accounting phenomenon and other accounting
phenomena were rated or performed on given dimensions
2. Distinctiveness information: how this accounting phenomenon was rated or
performed on various other dimensions
3. Consistency information: how this accounting phenomenon was rated or
performed on important dimensions in the past
Evidence shows that subjects tend to focus more on distinctiveness or consistency
information than on consensus information.54 Studies examining search behavior in
reaction to an accounting phenomenon are very limited.

The search behavior is not misguided. It is fair to assume that the decision maker has
some expectations about the accounting phenomenon that may de
Cognitive Relativism

termine the type of information sought. These expectations are termed preconceived
notions in A. S. De Nisi et al.s model.55 They result from the decision makers
previous experiences with the accounting phenomenon. These expectations or
preconceived notions may bias the decision maker toward choosing some information
rather than other information. Providing background information prior to observation
contributes to this phenomenon.56-57 R. S. Wyer and T. K. Srull maintain that prior
information predisposes the subject to select one of a number of frames of references. 58
Bias is a result of the tendency to seek evidence confirming preconceived notions
rather than neutral or disconfirming evidence.59-61
Schema Formation or Building
Once the accounting phenomenon has been observed, the relevant information is
encoded in the sense that it is categorized on the basis of experience and organized in
memory along schemata or knowledge structures. As put by R. E. Nisbett and L. Ross:
Few, if any, stimuli are approached for the first time by the adult. Instead, they are
processed through pre-existing systems of schematized and abstracted knowledgebeliefs, theories, propositions and schemas. These knowledge structures label and
categorize objects and events quickly and, for the most part, accurately. They also
define a set of expectations about objects and events and suggest appropriate responses
to them.62
A schema can be simply an update of templates that existed prior to the occurrence of
a known accounting phenomenon or a new template generated by the occurrence of a
new accounting phenomenon. In the first case, little ambiguity is assumed to exist and
therefore the encoding follows an automatic proc- ess.63 In the second case, no
immediate available schema exists, and a controlled categorization process is triggered
to determine which schema is consistent with the dimensions of the accounting
phenomenon. Both processes are suggested in the case of the encoding of information
or performance appraisal:
Thus, both the automatic and controlled processes have the same end result: the assignment of a person to a category based on prototype-matching process. The difference is
whether the stimulus persons behavior is sufficiently consistent with other cues to

allow the categorization to proceed automatically or whether a controlled process must

be used to determine which category is consistent with the individuals behavior. The
actual category assignment is a function of contextual factors influencing the salience of
particular categories and stimulus characteristics, as well as individual differences
among perceivers that tender some categories and their prototypes more available than
others and some stimulus features more salient than others.64
Basically, an accounting phenomenon may be categorized in a given schema, by
virtue of its possession of obvious or salient attributes known to the perceiver.
When no salient category prototype or schema provides a natural framework, the
automatic process is superseded by a controlled process or a consciously monitored
process.65
The controlled process can be triggered by either a new accounting phenomenon or
new features of a known phenomenon that are inconsistent with a previous
categorization. In the latter case a recategorization is invoked until the inconsistency is
resolved and a new schema is used to describe the accounting phenomenon, causing a
reconstruction of memories about the phenomenon such that memories consistent with
the new categorization are more available.
Schema Organization and Storage
After information about a given phenomenon is encoded to form a representation or
schema, it is stored and maintained in long-term memory. E. Tulving distinguishes
between episodic and semantic memory.66 Basically, a persons episodic memories are
personal while semantic memory is knowledge of words and symbols, their meanings
and referent knowledge of the relations among words, and the rules or algorithms for
manipulating words, symbols, and the relations among them. R. C. Atkinson and R. M.
Shiffrin maintain that the basic structural features of episodic memory are three
memory stores: the sensory register, the short-term store, and the long-term store. 67
Information enters the memory system through the various senses and goes first to the
sensory register whose function is to preserve incoming information long enough for it
to be selectively transmitted into the memory system. It is kept there less than a second
and is lost through either decay or erasure by overwriting.
The information then goes to the short-term store, working in memory where

conscious mental processes are performed. It is where consciousness exercises its

function. Information can be kept indefinitely here provided it is given constant
attention; if not, it is lost through decay in twenty to thirty seconds.
The information next goes to the long-term store through a conscious or unconscious
process where it can be held indefinitely and often permanently (although it can be lost
owing to decay or interference of various sorts). The long-term store is assumed to have
unlimited capacity. In this multistore model information about the accounting
phenomenon moves through different and separate memory systems, ending with a
long-term store where semantic information is maintained along meaning-based codes
or schemata. It is important to realize at this stage that if the person intends to
remember the accounting phenomenon for all time, he/she must perform a different
analysis on the input than when his/her intentions are temporary.68 A persons intention
determines whether the storage of the information on the accounting phenomenon is
permanent or temporary. A different coding is used: a memory code for permanent
storage and a perceptual role for temporary storage.
Different codes have different permanence. Codes of the sensory aspects of an input,
such as appearance, are short lived. Hence, a person who looked at a word to decide
whether it was printed in red or green would not remember the words name very long
because his coding would have emphasized color, not meaning. In contrast, a person
who looked at a word to decide whether it was a synonym for some other word would
form a semantic code, and he/she would remember the name of the examined word for
quite a while.69

Stimulus and Attention and Recognition Processes

Upon observation of a triggering event or stimulus, the schema in the accounting
phenomenon is activated. The activation, as a process of detection, search, and
attention, can be either a controlled or an automatic processing.70
Basically, automatic detection, triggered by the recognition of a stimulus, operates
independently of the persons control. Automatic processing is the apprehension of
stimuli by the use of previously learned routines that are in the long-term storage.
Automatic processing as learned in long-term store, is triggered by appropriate inputs,

and then operates independently of the subjects control. An automatic sequence can
contain components that control information flow, attract attention, or govern overt responses. Automatic sequences do not require attention, though they may attract it if
training is appropriate, and they do not use up short-term capacity. They are learned
following the earlier use of controlled processing that links the same nodes in sequence.
In search, detection, and attention tasks, automatic detection develops when stimuli are
consistently mapped to responses; then the targets develop the ability to attract attention
and initiate responses automatically, immediately, and regardless of other inputs or
memory load.71
In these automatic processes, no conscious effort is involved in the search as well as in
demanding attention owing to the learned sequence of the elements composing the
schemata. On the other hand, controlled processes involve a temporary activation of
novel sequences of processing steps that require attention, use short-term memory, and
involve a conscious effort.
It is important to realize that in both processes, the use of schemata for encoding or
retrieving information depends on accessibility in memory, where the accessibility of
schemata is the probability that they can be activated, either for use in storage of
incoming information or for retrieval of previously stored information.72-73
Accessibility of a schema depends upon such factors as the strength of the stored information, the extent of the overlap or match between input and schema, and the recency
and frequency of previous activations. Each time a schema is activated for use, it
becomes more accessible for successive activations. The instrumental effect of an
activation on the accessibility of a schema is presumably a decreasing function of its
prior strength. That is, a weak schema benefits more from an activation than a strong
one.74
Empirical evidence on the increased accessibility of information with the frequency
of activation is available.75,76
Retrieval of Stored Information Needed for Judgment/Decision
Either the automatic or controlled search processes activate the appropriate schema
for the accounting phenomenon and allow the retrieval of information on the

phenomenon. It is, however, the schema, a representation of the phenomenon, that is

recalled rather than the actual phenomenon.77-79 The effect becomes stronger as the time
between observation and recall increases.80
The potential for different types of biases exists at this stage. For example, people
may be more likely to recall information consistent with a schema confirming an
expectation,81 or may recall schema-consistent information that they never saw.82 A
good deal of evidence also suggests that schema-inconsistent information is more likely
to be recalled83 because of its novelty, saliency, and difficulty of incorporation into a
schema.84
What is more likely to be recalled when faced with an accounting phenomenon, what
types of biases affect the recall of schemata of accounting phenomena, and what can be
done to reduce or eliminate the distortions in recall are some of the important questions
in need of investigation. This model will assume that familiarity with the accounting
phenomenon through constant record keeping and other forms of monitoring may result
in less biased recall. The solution, in fact, is more complex and depends on the type of
relationship between memory and judgment. Reid Hastie and Bernadette Park
investigated these relationships and distinguished between two types of judgment tasks,
memory-based and on-line. They also identified five information-processing models
that relate memory for evidence to judgment based on the evidence: (1) independent
processing, (2) availability, (3) biased retrieval, (4) biased encoding, and (5)
incongruity-biased encoding.85
With regard to the five information-processing models, the distinction is threefold:
(1) cases where there is no relationship between judgment and memory processes,
which include the independent processing model; (2) cases where memory availability
causes judgment, which include the availability-based information-processing model
and the automatic search process described earlier; and (3) cases where judgment
causes memory, which include the biased retrieval, the biased encoding, and
incongruity-biased encoding models. The biased retrieval model is selective in the
sense that traces that fit the judgment are more likely to be found at the memory
decision stage. Such biases have been termed selective recall, confirmatory memory,
and access-biased memory.86-89
The biased encoding model assumes that biasing takes place at the time of the

encoding of evidence information and memory search will locate a biased sample of
information reflecting the initial encoding bias.
The incongruity-based encoding model assumes that after the initial encoding,
incoming information that is incongruent or contradictory is given special processing to
enhance its memorability by being placed in special tags that strongly attach to
memory. In memory search, the subject is more likely to find the incongruent
information.90-91 This model assumes that where the accounting phenomenon calls for
an online task, the availability or automatic search model will characterize the retrieval
of stored information needed for judgment decision. Selection of a processing model
will depend on the individual objectives of the subject and the perceived consequences
of his/her judgments on his/her economic and psychological welfare.
Reconsideration and Integration of Retrieved Information with Other
Available Information
At this stage the process involves integration of the information retrieved from
memory and other available information into a single evaluation of the accounting
phenomenon.
Where familiarity with the phenomenon is present and previously learned routines are
retrieved, active integration will not take place. An earlier integration is recalled from
past stored output on the phenomenon. What was once accomplished by slow,
conscious, deductive reasoning is now arrived at by fast, unconscious perceptual
processing.92 Where the phenomenon presents challenging and novel dimensions and
where controlled processes were involved in attention and recognition, a cognitive integration of all the information is required to reach a single evaluation of the accounting
phenomenon. G. Mandler describes the process of response learning as follows:
First, the organism makes a series of discrete responses, often interrupted by incorrect
ones. However, once errors are dropped out and the sequence of behavior becomes relatively stableas in running a maze, speaking a word, reproducing a visual pattern
the various components of the total behavior required in the situation are integrated.
Integration refers to the fact that previously discrete parts of a sequence come to behave
functionally as a unit; the whole sequence is elicited as a unit and behaves as a single

component response has in the past; any part of it elicits the whole sequence.93
Brunswicks lens model and Andersons weighted average model provide support to
the types of integration of information that take place. 94 The integration process is,
however, also subject to various biases:
1. People may attach and give great weight to some type of information. For
example, evidence in the employee appraisal literature shows that negative
information has greater weight.95,96
2. There is evidence in both psychology and accounting of an underutilization or
underweighing of base rate or consensus information.97
3. There is ample evidence in psychology and accounting of the effect of various
heuristics involved in decisions on and about accounting phenomena. They
include (1) representativeness, (2) availability, (3) confirmation bias, (4)
anchoring and adjustment, (5) conjunction fallacy, (6) hindsight bias, (7) illusory
correlation and contingency judgments, (8) selective perception, (9) frequency,
(10) concrete information, (11) data presentation, (12) inconsistency, (13)
conservation, (14) nonlinear extrapolation, (15) law of small numbers, (16)
habit/rules of thumb, (17) best-guess strategy, (18) complexity in the
decision environment, (19) social pressures in the decision environment, (20)
consistency of information sources, (21) question format, (22) scale effects, (23)
wishful

thinking,

(24)

outcome-irrelevant

learning

structures,

(25)

misperceptions of chance fluctuations (gamblers fallacy), (26) success/failure

attributions, and (27) logical fallacies in recall.98
The Judgment Process
The judgment process is the result of the integration process of information and the
forming of a single evaluation of the accounting phenomenon if the attention,
recognition, and integration processes are the result of controlled processes. The
judgment made in this case requires a conscious access to all the mental processes
implied in the model. If, however, the attention, retrieval recognition, and integration
processes were the result of automatic processes, the judgment is not and will not be
conscious. It does not require the conscious use of all the mental processes implied in

this model.99-100 It is a routine judgment.

Routine judgment involves the rapid matching of immediate perceptions to a
template that provides, and executes, a specific response: if total debts do not equal
total credits, readd the total balance.
In the above example, there is no awareness of how the brain actually decides that the
debits do not equal the credits. Even if awareness were possible, it is not normally
necessarya great many of our routine activities, such as keeping our eyes open or
holding our pencilsare done without any particular conscious awareness, at least until
something causes us to become aware.101

Decision/Action (Response)
The final step of the model is the decision or selection of a response to the accounting
phenomenon. It is a conscious response preference resulting from the judgment process.
It is an output of the judgment process and is clearly influenced by all the mental
processes and biases described earlier. As a result, a new schema on the phenomenon
will develop that will be part of the knowledge structure or the phenomenon stored in
long-term memory.
The move from judgment to decision is a bridging process. It assumes that no
obstacles stand in the way.
The decision/action has been investigated in various accounting environments and
using various accounting phenomena. It has been found to differ from various
normative decision models, including Bayerian-decision theory and expected value
models.102-103
The bridging process, however, will be influenced by the cognitive steps described
in this model as well as by other factors including the possible consequences of the
decision on the accounting phenomenon. Gibbins, for instance, cites the following
factors:
Personal attitudes may play a direct role, such as determining priorities within the
search process. For example, some public accountants may use financial return as their
first selection criterion; others may use moral propriety as their first. Personal attitudes
can also play an indirect role, limiting past actions and thus limiting the experiences on

which judgment guides are built. The applications of such attitudes to the judgment
process need not be consciousparticularly for deeply ingrained beliefs.104
CONCLUSIONS
The essence of cognitive relativism in management accounting is the presence of a
cognitive process that is assumed to guide the judgment/decision process. The model in
this chapter shows that judgments and decisions made about accounting phenomena are
the products of a set of social cognitive operations that include the observation of
information on accounting phenomena and the formation of schemata that are stored in
memory and later retrieved to allow the formation of judgments and/or decisions when
needed.