GENERELIZATION AND

MAINTAINANCE IN ADU
LT LANGUAGE DISORD
ERS
 GENERILIZATION
• focuses on the efficiency of the treatment
• The influence of the intervention on other untrained behaviours
that may or may not have some type of systematic relation to the
trained behaviour

MAINTAINENCE
• The stability of an acquired beavioural change
over time in the absence of continued
intervention
 GENERALIZATION

Generalization is progress beyond the treatment, either in other

conditions or with untrained responses.

In single- case experiments, investigators look for generalization that

is related to a treatment activity, such as word- finding treatment with
naming and generalization to words used in conversation

The common ground between the therapy and the measure is the
process being treated (i.e., word retrieval)
There are 3 type of generalization

• stimulus generalization
• Response generalization
• Maintenance
Response generalisation is the emergence of an untrained language response
following intervention

stimulus generalisation is the transfer of a trained behaviour to another

context.
In therapy methods for creating generalization, there are two types:

•Within level generalization

•Across level generalization

Within level generalization

Within-level generalisation is defined as change to untreated stimuli

within the same linguistic level as the focus of treatment

e.g., generalisation from treated words to untreated words in a single-

word naming task, generalisation from treated sentence structures to
untreated structures in a sentence production task.

Within-level generalisation will normally be response generalisation

as the elicitation stimulus is generally the same
Across-level generalisation

Across-level generalisation is defined as change at a linguistic

level different to the focus of treatment,

e.g., production of words order level in sentences, production of

treated/untreated syntactic structures in connected speech
following sentence-level therapy.

The change across levels generally involves a change of stimulus

or condition so it will be stimulus generalisation
The below -mentioned framework is used to review the current evidence
regarding therapies for spoken production, considering within- and across-level
generalisation and exploring single-word therapies, sentence production,
and discourse therapies

Two disinctions are drawn here. The first is between

(i) single-level treatment—treatments that target a single linguistic level
(ii) multilevel treatment—treatments that target more than one level. The
second is between studies that measure within-level generalisation only
and those that measure both within- and across-level
generalisation.
Word level

Within level :
participants have been taught to use access to intact orthographic
information to support spoken production, including teaching grapheme-
to-phoneme conversion to produce self-generated phonemic cues, use of
cueing aids or use of a letter board

Based on a study it has shown that 13 people out of 19 was able to

achieve generalization through this technique.

within-level generalisation at the single-word level is influenced by

treatment approach, the focus of therapy, and the underlying impairment
Across level:

There has been greater consideration of across-level generalisation in studies of

verb retrieval, possibly reflecting the integral role that verbs play in sentence
production.

Studies have investigated gains at sentence level and/or in connected speech.

Within-level gains in the retrieval of treated verbs have been accompanied by
improvement in sentence production around treated verbs More general gains are
also seen around untreated verbs or on general tests of sentence production.

Gains appear to be independent of the nature of therapy and the underlying verb
deficit although cooccurring syntactic deficits can restrict improvement
Sentence level
the two therapy techniques that can be used :
verb-centred mapping, Treatment of Underlying Forms

Within level:
Verb-centred mapping therapies focus on identifying the verb and
other sentence components in written sentences, with the aim of
increasing awareness of sentence structure and thereby improving
production

TUF involves the identification of sentence components in an active

sentence, followed by instructions about how to manipulate the
constituents to produce the complex form.

Within level, some generalisation has been seen across sentence types
following both therapies
Across-level generalisation.:

In therapies that have focused on sentence production, across-level

gains have been reported in connected speech, primarily narratives with
therapy focused on comprehension resulting in gains in production. The
mapping therapies have resulted in an increased use of verbs and in
an increase in verb argument structures
Discourse level

Therapy involved repeated practice of the trained discourse task, with the
person problem solving and developing strategies to achieve effective
communication.

but there was no generalisation to performance on the Communication

Activities of Daily Living-2 which involves other everyday scenarios;
generalisation to related discourse tasks was therefore not seen.
Multi level therapies

The treatments aim to maximise change by providing a more natural

context for therapy and/or promoting generalisation to everyday
communication. The section is divided into different types of
studies:
(i) word and sentence
(ii) word and connected speech
(iii) sentence and discours
(iv) word, sentence, and discourse.
Word- and sentence-level therapies

It focuses on therapy techniques that combines verbs and sentence

production

It can include
•treating verbs within a sentence context

•combining work on verb retrieval sentence production

•aiming to increase awarness of verb argument structure by focusing

on verbs and their associated
word- and connected-speech-level therapies

The use of more natural everyday contexts may be needed to

promote generalisation of the use of targeted words in connected
speech but may not be sufficient (without structured repetitive tasks)
to produce specific gains in word retrieval.
Word-, sentence-, and discourse-level therapies

Milman, Vega-Mendoza, and Clendenen (2014) used a multilevel approach,

Integrated Training for Aphasia (ITA), with three participants with aphasia.

ITA involved the delivery of word-, sentence-, and discourse-level therapy (20
min each of a 1-hr session) where the single vocabulary items were incorporated
into the sentence and discourse components, and the morphosyntactic structures
included in discourse.

The targets , were words and sentences rather than any direct focus on
discourse features.
Within-level generalisation was not found at the word level for
semantically related words and, across level, the relationship
between change at word and sentence levels was not clear.

Findings from this study showed that within-level

generalisation was seen at both the word and discourse levels,
 Factors that promote generalization of
learned
from trained to untrained
This generalization is only observed some of the time, and we are not yet
certain of the factors that promote this generalization. Some factors
thought to affect generalization are:
 Linguistic relatedness among items being trained

 Prototypicality of items in a category being trained

 The type of learning that a particular treatment approach is following

 Generalization probe

 A generalization probe is administered at the end of a session

 These post session probes may be given daily, on alternate days, or

once per week Reliability is maximized by using at least 10 items and,
of course, by administering the task the same way each time

 A generalization probe is linked to a treatment activity by a goal as

well as a process.

 Sometimes the goal is long- term, such as improving natural

conversation while the therapy is still focused and simple
 the generalization probe tells us whether we are meeting the goal.

 The degree of difference between a treatment task and a generalization

probe varies as a function of whether a goal is short- term or long- term

 This table shows us the relationship between goal treatment and

generalization
Generalization of therapy effects on sentence
processing skills

 Treatment approaches demonstrate that aphasic sentence processing

skills responds positively to therapy

 Despite these positive findings, the degree of change varies from

study to study, particularly in terms of generalization

 Clinically, evidence of generalization enhances the claim that

therapy is of functional benefit.

 Theoretically, generalization can help to identify processing changes

that may have occurred as a result of our intervention
MAINTAINANCE

Many aphasia treatments have been shown to be successful, but decline

from post-treatment levels at 3 months ,1 month and even 1 week

Maintenance is rarely tested at longer intervals.

To say that a treatment method to be considered successful there should be

good maintenance

Follow up are occasionally done after aphasia treatment the maintenance

of treatment goals are rarely checked

Different studies were done to understand which all methods helps in

maintaining the treatment goals for a long term
To understand the maintenance a study done by Elizabeth H. Lacey,

 The study focuses on the effect of error-production during treatment on

initial learning and long-term maintenance of a behavioral intervention
for word-finding (anomia)

 They introduced it with 2 methods of teaching i.e

error less method of teaching
error full method of teaching

 Errorful (EF) version in which guessing is encouraged and errors

occur frequently

 Errorless (EL) version in which supportive cues greatly reduce the

chances of making an error
 Errorless (EL) version in which supportive cues greatly reduce the
chances of making an error

 They hypothesized that EL treatment would be associated with better

maintenance, especially in patients with memory deficits, but the
comparison revealed no advantage to EL.

 The results suggest that attentional engagement may be more relevant

than error-prevention in the rehabilitation of people with aphasia.

 Many researches and studies are conducted in order to find the treatment
methods which provides long term maintenance for patients with
language disorders
BIBLIOGRAPHY
• Ilias Papathanasiou, Patrick Coppens, Constantin Potagas, Aphasia
and related neurogenic disorders , Jones and Bartlett
learning,2013,MN city

• G.Albyn Davis, Aphasia and related cognitive communicative

disorders, Pearson education Inc., 2014, Boston

• Argye E Hillis, The handbook of adult language disorders 2nd

edition, Psychology press, 2015, new York

• Elizabeth H Lacey,Generalization and maintenance in aphasia

rehabilitation,, 2010, Washington D.C
Thank you
SUBMITTED BY
ANEENA K BENOY
ANGELA BINU