Neuropsychologia 46 (2008) 1940–1953

Phonological short-term store impairment after cerebellar
lesion: A single case study
Francesca R. Chiricozzi a,b , Silvia Clausi a,b , Marco Molinari b , Maria G. Leggio a,b,∗
a Department of Psychology, University of Rome “La Sapienza”, Rome, Italy
b I.R.C.C.S. Santa Lucia Foundation, Ataxia Lab, Rome, Italy

Received 26 July 2007; received in revised form 17 January 2008; accepted 25 January 2008
Available online 7 February 2008

Abstract
The cerebellum is a recent addition to the growing list of cerebral areas involved in the multifaceted structural system that sustains verbal
working memory (vWM), but its contribution is still a matter of debate. Here, we present a patient with a selective deficit of vWM resulting from a
bilateral cerebellar ischemic lesion. After this acute event, the patient had impaired immediate and delayed word-serial recall and auditory-verbal
delayed recognition. The digit span, however, was completely preserved. To investigate the cerebellar contribution to vWM, four experiments
addressing the function of different vWM phonological loop components were performed 18 months after the lesion, and results were compared
with normative data or, when needed, with a small group of matched controls.
In Experiment 1, digit span was assessed with different presentation and response modalities using lists of digits of varying lengths. In Experiment
2, the articulatory rehearsal system was analyzed by measurement of word length and articulatory suppression effects. Experiment 3 was devoted to
analyzing the phonological short-term store (ph-STS) by the recency effect, the phonological similarity effect, short-term forgetting, and unattended
speech. Data suggested a possible key role of the semantic component of the processed material, which was tested in Experiment 4, in which word
and nonword-serial recall with or without interpolating activity were analyzed.
The patient showed noticeably reduced scores in the tasks that primarily or exclusively engaged activity of the ph-STS, namely those of
Experiment 3, and good performance in the tests that investigated the recirculation of verbal information. This pattern of results implicates the
ph-STS as the cognitive locus of the patient’s deficit. This report demonstrates a cerebellar role in encoding and/or strengthening the phonological
traces in vWM.
© 2008 Elsevier Ltd. All rights reserved.
Keywords: Working memory; Cerebellum; Verbal span; Lexical–semantic coding

1. Introduction
Working memory is a form of temporary storage of information that is needed for performing a wide range of cognitive
skills (Baddeley, 2003). This mechanism allows the temporary
maintenance and manipulation of a limited amount of information. According to the Baddeley–Hitch model (Baddeley &
Hitch, 1974), the working memory system does not have a unitary functional architecture but comprises a limited capacity
attentional controller (central executive) and two subsystems

∗ Corresponding author at: Head Ataxia Lab Santa Lucia Foundation, Department of Psychology, University of Rome “La Sapienza”, Via dei Marsi 78, 00185
Roma, Italy. Tel.: +39 0649917645; fax: +39 0649917711.
E-mail address: maria.leggio@uniroma1.it (M.G. Leggio).

0028-3932/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuropsychologia.2008.01.024

that temporarily store visual-spatial information (visuospatial sketchpad) or acoustic-verbal information (phonological
loop).
The phonological loop is further divided into two independent components: the phonological short-term store (ph-STS),
which is involved in immediate retention of verbal information;
and the rehearsal system, which refreshes phonological information to prevent its rapid decay (Salamè & Baddeley, 1982).
If verbal material is presented in spoken form, access to the
ph-STS is direct and obligatory. If verbal material is presented
visually (through written stimuli or pictures), then it needs to be
recoded into phonological code by means of the rehearsal process before accessing the ph-STS (Levy, 1971; Sperling, 1963).
Thus, the rehearsal process has two functions: to refresh phonological traces and to translate visual stimuli into a phonological
code.

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

Recirculation of the memory trace, provided by the rehearsal
system, is supposed to take place between the ph-STS and the
so-called phonological output buffer (Vallar, Basso, & Bottini,
1990), a working memory space in which phonological segments are temporarily stored to allow the programming of
various output processes (Burani, Vallar, & Bottini, 1991). Evidence from various fields of neuroscience demonstrates that the
working memory system has discrete neural correlates within
the left frontal and parietal cortical areas (Paulesu, Frith, &
Frackowiak, 1993; Silveri & Cappa, 2003; Vallar, Di Betta,
& Silveri, 1997; Vallar & Papagno, 1995). Moreover, some
data indicate that subcortical structures are also involved in
working memory functions. Specifically, neuroimaging studies
have demonstrated that the cerebellum is one of the most consistently activated regions within a network that includes the
inferior frontal lobe, the supplementary motor area, the premotor cortex, and the parietal lobe (Andreasen et al., 1995; Chen
& Desmond, 2005b; Chein & Fiez, 2001; Desmond, Gabrieli,
Wagner, Ginier, & Glover, 1997; Grasby et al., 1994; Kirschen,
Chen, & Schraedley-Desmond, 2005; Li et al., 2004; Paulesu et
al., 1993).
In addition to neuroimaging data, lesion studies demonstrating mild impairment of verbal working memory (vWM) tasks
in patients affected by cerebellar lesions provide further support
for a cerebellar role in working memory. (Justus, Ravizza, Fiez,
& Ivry, 2005; Ravizza, McCormick, Schlerf, Ivry, & Fiez, 2006;
Silveri, Di Betta, Filippini, Leggio, & Molinari, 1998).
Consistent with the proposals advanced in Silveri et al.’s
(1998) neuropsychological study, most neuroimaging studies
hypothesise a cerebellar role in the rehearsal process, arguing
that the cerebellum, in conjunction with the inferior frontal lobe
and the supplementary motor area, is part of a circuit that sustains
articulatory rehearsal mechanisms, whereas the inferior parietal
lobe is the locus of the ph-STS (Chein & Fiez, 2001; Grasby
et al., 1994; Paulesu et al., 1993; Petrides, Alivisatos, Meyer, &
Evans, 1993). This assumption is consistent with the hypothesis
that the cerebellum controls speech production at the interface
of the subvocal rehearsal process and motor execution (Silveri
et al., 1998).
However, articulatory rehearsal does not seem to provide
a complete account of cerebellar involvement in verbal working memory. Significant activation is still observed in the right
cerebellar hemisphere when the rehearsal control condition is
subtracted from the working memory task (Awh et al., 1996),
suggesting that the cerebellum performs functions in addition
to, or instead of, articulatory rehearsal. Contrary to the rehearsal
hypothesis, Chein and Fiez (2001) found no cerebellar activation
during the maintenance phase of a delayed recall task; however,
they described primary activation during encoding and retrieval
working memory phases.
Lesion studies have also questioned the notion that the cerebellar role is confined exclusively to rehearsal mechanisms. The
requirement to speak while trying to remember verbal material in a working memory task is detrimental to recall, because
articulatory mechanisms are unavailable for rehearsal (articulatory suppression) (Baddeley, Thomson, & Buchanan, 1975).
Unlike patients with prefrontal cortex lesions (Goerlich, Daum,

1941

Hertrich, & Ackermann, 1995; Vallar et al., 1997), patients with
cerebellar damage show normal suppression effects (Ravizza et
al., 2006; Silveri et al., 1998).
This evidence questions the selective influence of the cerebellum on rehearsal and supports the idea proposed by Ravizza
et al. (2006) of a more complex cerebellar contribution to both
storage and rehearsal working memory mechanisms. According to this hypothesis, two different cerebro-cerebellar networks
– i.e., one for articulatory rehearsal and one for phonological storage – have been described (Chen & Desmond, 2005b).
Articulatory rehearsal refers to the frontal-superior cerebellar
articulatory control system, in which the superior cerebellar
hemisphere receives input from frontal areas via the medial pontine nuclei; phonological storage refers to the parietal-inferior
cerebellar system in which the inferior cerebellar hemisphere
receives input from parietal areas involved in the ph-STS via the
lateral pontine nuclei (Chen & Desmond, 2005b). Until now,
however, no evidence of selective damage of the ph-STS with
preservation of rehearsal mechanisms in patients with cerebellar
damage has been demonstrated.
To detail the cerebellar contribution to verbal working
memory, the function of different vWM phonological loop
components was addressed in a patient with vascular bilateral
cerebellar lesions.
2. Case description
BB, a 54-year-old, right-handed male banker with 18 years
of formal education, was admitted to the Neurology Ward
of the Catholic University of Rome in August 2002. His
symptoms included the sudden appearance of tinnitus, sensory abnormalities on the left side of the face, and dizziness
with nausea and vomiting. The patient was alert and oriented
in time and place; a physical examination revealed ataxia of
gait, dysmetria of the left limbs, and horizontal torsional nystagmus.
MRI showed two T1 hyperintense areas in the left posteroinferior cerebellar hemisphere (lobule VIIIA) and in the right
antero-superior cerebellar regions (lobule V), confirming the
diagnosis of ischemic stroke. The extent of the lesions was
characterized by lobular parcellation according to Schmahmann,
Doyon, Toga, & Petrides’s (2000) atlas (Fig. 1). All extracerebellar structures were completely spared.
After 9 days, the patient was moved to Santa Lucia
Rehabilitation Hospital in Rome, at which time a general neuropsychological examination was performed. It showed that
intelligence (Raven, 1947), constructional praxia (Rey, 1968),
and frontal lobe abilities (Caltagirone et al., 1995) were within
the normal range. Spontaneous speech was spared. Verbal and
visuospatial spans (Orsini et al., 1987) were normal, as were
visual immediate recall abilities (Rey, 1968).
A test of memory and learning (Reynolds & Bigler, 1995)
revealed impairment of short-term and delayed auditory free
recall, as well as divergent results on delayed recognition. In the
latter test, the patient presented with a disproportionate deficit
in the auditory-verbal modality, with normal performance in the
visual-verbal modality. No further testing was carried out at that

1942

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

Fig. 1. T1-weighted not-normalized MRI and T1-weighted MPRAGE-normalized SPM2 (Welcome Department of Cognitive Neurology, London, UK) images
showing ischemic lesions localized in the right lobule V (A) and in the left lobule VIIIA (B). Lobules are defined according to Schmahmann, Doyon, Toga, and
Petrides (Schmahmann et al., 2000).

time. The patient’s scores, adjusted for age and education, are
summarised in Table 1.
3. General neuropsychological assessment
BB came under our observation 18 months after the stroke.
At that time, his neurological examination revealed only slight
gait ataxia. His cognitive level and neuropsychological profile
were investigated. Four normal control subjects, matched for age
(mean age: 54 years) and education (18 years), were also tested;
the neuropsychological profiles of BB and the control subjects
are reported in Table 2.
Table 1
Neuropsychological examination at time of lesion
Tasks

Cut-off

Cognitive profile
Raven-coloured matrices

28.6

18.96

Memory tasks
Span
Digit forward
Digit backward
Spatial forward
Spatial backward

6
5
6
5

7±2
5±2
7±2
5±2

Visuospatial memory
Rey’s complex figure

24

22 ± 4.9a

Words serial recall (TOMAL)
Free immediate recall
Free delayed recall
Verbal cues recognition
Visual cues recognition

5
5
0
9

7
7

Constructional praxia tasks
Rey’s complex figure

33

32 ± 1.8a

Executive tasks
Analogy

18

18

a Mean raw score values from Rey (1968).

BB’s total, verbal, and performance IQs were within the normal range according to the WAIS-R (Wechsler, 1997), and his
linguistic and praxic abilities were intact as measured by the
phonemic word fluency task (Borkowsky, Benton, & Spreen,
1967) and by “Mental Deterioration Battery” (Caltagirone et
al., 1995). His performance on the immediate visual recognition
test was similar to those of the controls (Caltagirone et al., 1995).
He made a few omissions on the double line cancellation task
(Zazzo, 1980), suggesting a mild defect in visual exploration
ability.
BB’s performance on the word-serial recall task, extrapolated from TOMAL (Reynolds & Bigler, 1995), showed that he
had clearly recovered from the acute phase. In immediate and
delayed free recall, BB recalled 11 and 12 of 12 items, respectively, and thus, delayed recognition was not tested (Table 2).
Spatial and verbal span were measured using Corsi’s blocktapping test (Orsini et al., 1987) and immediate repetition
of digits (Orsini et al., 1987), words, and nonwords (Miceli,
Laudanna, & Burani, 1994). BB’s scores were normal for
block-tapping and repetition of digits and words. However,
the patient’s performance of nonword stimuli retrieval was
impaired, exhibiting clear differences from controls. When
manuality-handedness was assessed by the Edinburgh Inventory (Oldfield, 1971), the patient obtained a score of 10/10 in
favour of his right hand.
In summary, the patient’s neuropsychological profile 18
months after the acute event confirmed intact praxis and visuospatial and verbal fluency abilities, as reported in the acute
phase. A mild, previously undiscovered deficit in visual exploration was present. The patient’s performance on the word-serial
recall task showed recovery with scores comparable with those
of controls.
His verbal span task performance, however, was intriguing.
In fact, while BB’s performance was completely preserved on
the digit and word span tasks, his nonword span score was
clearly impaired. This observation, indicating a possible defect
in vWM, propelled us to further investigate the function of vWM
components.

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

1943

Table 2
Neuropsychological examination at 18 months after stroke
Tasks

Controls

Cut-off

BB

C1

C2

C3

C4

Cognitive profile
WAIS-R
Total IQ
Verbal IQ
Performance IQ

116
119
108

116
106
127

130
125
130

124
122
124

128
123
130

Memory tasks
Span
Digit forward
Digit backward
Spatial forward
Spatial backward
Word
Nonword

6
5
6
5
4
2

6
4
6
4
4
4

6
5
6
5
4
4

7
5
6
5
4
3

6
4
5
4
4
3

7±2
5±2
7±2
5±2

18.4

19.2

20.4

19.4

20.6

13.85

11
12

6
10
1
1

11
11
1

8
11
1

9
10
2

7
7

a

a

a

Visual recognition
Immediate picture recognition
Words serial recall (TOMAL)
Free immediate recall
Free delayed recall
Verbal cues recognition
Visual cues recognition

a
a

Visuospatial tasks
Double line cancellation

9

10

13

11

13

12

Constructional praxia tasks
Copying drawings
Copying drawings with landmarks

9.9
67.4

8.9
64.4

10
64.6

11.9
68.6

11.1
68.8

7.18
61.85

Linguistic tasks
Word fluency

41.6

28.2

37.6

41.6

40.2

17.35

a Not tested according to TOMAL protocol. Test is not performed if in the preceding recall tasks 12 out of 12 words are already recalled.

4. Experiment 1: digit span
Evident or mild digit span deficits have been reported in
cerebellar patients (Maddox, Aparicio, Marchant, & Ivry, 2005;
Ravizza & Ivry, 2001; Silveri et al., 1998; Witt, Nuhsman, &
Deuschl, 2002). Digit span is the traditional method for measuring phonological short-term memory capacity (Vallar et al.,
1990; Vallar & Papagno, 1995), consisting of immediate serial
recall of digit sequences of increasing length. The score is the
length of the longest correctly recalled sequence. In verbal serial
recall tasks, stimuli can be administered in the auditory modality (wherein digits are spoken by the examiner) or in the visual
modality (by presenting a series of black digits printed on white
cards).
Independent of the presentation condition, two recall modalities can be used: verbal repetition and pointing recognition.
When a repetition response is required, the subject has to repeat
the stimuli verbally (auditory-verbal or visual-verbal). When a
pointing response is required, the subject has to indicate the
proposed sequence among alternatives (auditory-pointing or
visual-pointing). According to different presentation or recall
modalities, different components of the phonological loop can
be tested. In fact, while auditory input has direct obligatory

access to the ph-STS, visual material requires two additional
operations: phonological recoding (grapheme-to-phoneme conversion) and articulatory rehearsal.
Thus, patients with a ph-STS deficit present with a disproportionate impairment of the auditory span compared with the
visual span, and response modality does not affect performance
(Vallar et al., 1997).
Conversely, since verbal visual stimuli must be recoded by
the rehearsal system into a phonological code to gain access
to the ph-STS (Shallice & Vallar, 1990), a deficit of articulatory rehearsal is indicated by improved performance in the
pointing procedure after verbal presentation of stimuli than after
visual presentation (Silveri et al., 1998). Furthermore, a comparison between verbal repetition and pointing recognition allows
assessment of the phonological output buffer, a component of
rehearsal (Vallar et al., 1997). The memory advantage of the
pointing modality over the verbal repetition modality indicates
a reduced capacity of the phonological output buffer (Vallar et
al., 1997).
In summary, different performance patterns can identify specific loci of impairment of the vWM system. Therefore, we can
identify: (i) the ph-STS impairment pattern, characterized by a
pathological auditory-verbal span, associated with an auditory-

1944

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

verbal span lower than the visual-verbal span, and lack of an
advantage with pointing responses; (ii) the rehearsal component defect pattern, characterized by a visual-verbal span lower
than the auditory-verbal span, with possible mild impairment
of the auditory-verbal span; (iii) the phonological output buffer
impairment pattern, characterized by an advantage with pointing
responses versus verbal responses in every input modality.
Digit span was assessed using lists of digits of various lengths
with different presentation and response modalities.

et al., 1992). No advantage was found for the auditory modality over the visual modality. Immediate repetition of digits was
completely correct for both auditory and visual presentation. A
slightly higher span was detected in the visual-pointing condition than in the auditory-pointing condition.
In Experiment 1, BB’s digit span was preserved. A slight
advantage emerged in the visual-pointing modality with respect
to all other conditions, demonstrating good phonological recoding. Thus, BB’s digit span data did not pinpoint the functional
locus of the deficit. In the next two experiments, we specifically
investigated the different components of the phonological loop.

4.1. Methods
4.1.1. Participants
BB’s digit span performance was compared with those of the controls
reported in Vallar, Corno, and Basso (1992). These controls were 12 normal
subjects with a mean age of 50.52 years (range 25–74) and a mean educational
level of 10.08 (range 4–17) years.

5. Experiment 2: articulatory rehearsal system
assessment
In Experiment 2, we addressed the function of the articulatory rehearsal system by analysing the word length effect and
articulatory suppression.
Since the articulatory rehearsal system is a time-limited loop
that recirculates information stored in the ph-STS, long words
require more time to be articulated and thus are recalled less
frequently than short words (Baddeley et al., 1975). This effect
is described as the word length effect (Baddeley et al., 1975). In
normal subjects, the word length effect is a good indicator of the
efficiency of the articulatory rehearsal system.
Articulatory suppression refers to the disruption of verbal
material retention in immediate memory span tasks by means
of continuous uttering of an irrelevant speech sound. This effect
is due to the interference of uttering on rehearsal activity and
is considered an indirect index of the rehearsal process function (Baddeley, Lewis, & Vallar, 1984; Levy, 1971; Murray,
1968; Peterson & Johnson, 1971). In the case of a nonoperative rehearsal process, no articulatory suppression is expected
(Silveri et al., 1998). Conversely, a strong suppression effect
indicates intact recirculation of verbal information (Vallar et al.,
1997).

4.1.2. Materials
The span task was administered using lists of digits in various presentation
and response modalities. The stimuli set consisted of 5 sequences of 1, 2, 3, 4,
5, 6, or 7 digits. No digit was presented twice within a sequence.
4.1.3. Design and procedure
The digit span task was administered under four different conditions: (1)
auditory-verbal (the patient repeated the sequence spelled out to him by the
examiner); (2) visual-verbal (the patient repeated the sequence of digits visually presented one-by-one by the examiner); (3) auditory-pointing (the patient
indicated the sequence of digits spelled out by the examiner among various alternatives printed on white cards randomly displayed on the table); (4)
visual-pointing (the patient indicated the sequence of digits visually presented
one-by-one by the examiner among various written alternatives). The digits
included in each series were given to the patient at a presentation rate of one digit
per second. If the patient did not correctly recall at least one of the sequences,
subsequent longer list was not administered.
The span score was computed by the “K + x” scoring system proposed by
Zhang and Simon (1985) and used by Vallar et al. (1992) in an Italian population
study. Sub-score K was the length of the longest sequence correctly recalled in
5 of 5 (100%) lists. Sub-score x was the number of correctly recalled sequences
1 item longer than K. Since each correctly recalled sequence was worth 0.20,
the x sub-score ranged from 0 to 0.80 (0 to 4 of 5 sequences correctly recalled).

5.1. Methods

4.2. Results

5.1.1. Participants
In addition to BB, four control subjects, matched to the patient for age (mean
age: 54 years) and education (18 years), were subjected to the word length effect
test.

BB’s performance, reported in Table 3, was normal compared with those reported in same-age control subjects (Vallar
Table 3
BB’s span performances in different conditions
Conditions

Auditory-verbal

Auditory-pointing

Visual-verbal

Visual-pointing

BB span
Controls (Vallar et al., 1992)

6.80
5.52

6.80
5.95

6.60
5.64

7
5.41

Sequence length

Percentage of correct responses

1
2
3
4
5
6
7

100
100
100
100
100
100
80

Span are computed following Vallar et al. (1992).

100
100
100
100
100
100
80

100
100
100
100
100
100
60

100
100
100
100
100
100
100

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953
5.1.2. Materials
5.1.2.1. Word length effect. To investigate the word length effect, a modified
version of the original test developed by Brizzolara, Casalini, Chilosi, and
Cipriani (2002) was administered. The adapted version consisted of two lists
of two- or four-syllable words (Lists 1 and 5 of the original version) of high-use
frequency, presented in either auditory or written form.
5.1.2.2. Articulatory suppression. Stimuli consisted of sequences of 7 digits
each. No digit was presented twice within a sequence. A total of 4 blocks of 20
series of items were given to the patient. The response set material comprised
10 digits (from 0 to 9) printed in random positions on a white card.
5.1.3. Design and procedure
5.1.3.1. Word length effect. Items for this test were organised in progressively
longer series (2–6 items). For each list (1–5), the examiner administered each
string of words at a rate of one per second. The subject had to repeat the items
in the order in which they were presented. If he succeeded, strings of words of
increasing length were presented. If the subject failed to repeat a string, a second
series of the same length was administered. If the subject failed again, the test
was discontinued.
The span was established as the length of the last correctly recalled string. In
the auditory presentation modality, the stimuli were read aloud by the examiner;
in the visual presentation modality, the strings of words were printed on white
cards and shown to the subject.
The patient’s two- and four-syllable word spans were compared by frequency
distribution test (χ2 -value).
5.1.3.2. Articulatory suppression. This task was articulated in two sessions. In
each session, BB received two blocks of stimuli presented under two different
conditions (I+ and I−) in an ABBA design. In the articulatory suppression
condition (I+), the patient was requested to continuously pronounce a nonsense
word (blah), starting before the examiner presented each auditory list of digits
and stopping after immediate recognition of the stimuli in the presentation order.
Recognition was executed by pointing to the digits of the white response board
(Baddeley et al., 1984). The pointing response was necessary because of the
continuous uttering of verbal speech throughout the task.
In the control condition (I−), the patient was not engaged in any concurrent
activity. According to Vallar et al. (1997), for articulatory suppression, one must
use a digit list length for which at least 70% correct recognition was scored in the
standard auditory-pointing span task. Thus, a list of 7 digits was administered to
BB (see Table 3). Correctly recognised sequences by BB under the suppression
and control conditions were compared by frequency distribution test (χ2 -value).

5.2. Results
5.2.1. Word length effect
The patient presented no length effect with auditory stimuli;
on the contrary, his performance on long words exceeded that
for short words, suggesting a paradoxical word length effect.
The word length effect was absent in two out of the four control
subjects; i.e., their scores did not depend on the length of the
presented stimuli. The remaining two controls presented a normal word length effect; i.e. their scores were better with shorter
than longer words. Conversely, both BB and control subjects
exhibited normal length effects with visual verbal stimuli. In
the visual verbal condition, BB’s span was significantly higher
for two-syllable words than for four-syllable words (χ2 = 9.86,
d.f. = 1, p = 0.00). The results for BB and control subjects are
reported in Table 4.
5.2.1.1. Articulatory suppression. BB correctly recognised 24
of 40 auditory sequences under the suppression condition (60%)

1945

and 33 of 40 under the control condition (82.5%) (χ2 = 4.94,
d.f. = 1, p = 0.02).
The presence of a word length effect with visual stimuli and
of noticeable articulatory suppression demonstrates good functioning of the rehearsal process. However, the lack of a word
length effect in the auditory modality may raise doubts about
this finding.
In any case, two of the control subjects did not perform better with longer auditory words; further, there are reports on the
absence of the word length effect in patients with preserved
rehearsal systems and impaired ph-STS (Silveri & Cappa, 2003;
Vallar et al., 1997). This finding has been interpreted as a strategic choice, because rehearsing the material did not improve
memory performance in the presence of a dysfunctional ph-STS
(Vallar et al., 1997).
Interestingly, BB not only failed to show any word length
effect in the auditory presentation, he also had a slight advantage
with longer words. Different authors (Silveri et al., 1998; Vallar
& Cappa, 1987; Vallar et al., 1997) have ascribed this paradoxical word length effect to the ‘trace decay plus rehearsal’ theory
(Baddeley, 1990; Baddeley & Hitch, 1974; Brown & Hulme,
1995; Schweickert, Guentert, & Hersberger, 1990; Stigler, Lee,
& Stevenson, 1986). According to this concept, a partially
decayed item can be recalled by reconstructing its phonological trace. This process may be length-sensitive—for example,
when a phonological segment is lost, reconstruction has to rely
on the remaining segments. Therefore, long items are expected
to be easier to reconstruct than short items.
6. Experiment 3: phonological short-term storage
assessment
In Experiment 3, we assessed the function of the ph-STS
by analysing recency and phonological similarity effects, shortterm forgetting, and unattended speech.
In immediate free or serial recall of auditory lists, normal
subjects typically present a U-shaped performance, with higher
retention of the initial items (primacy effect) and final items
(recency effect) than of the intermediate ones. Retention of the
initial and intermediate stimuli primarily engages the articulatory process, whereas recalling the final ones involves the
ph-STS in particular (Vallar & Papagno, 1986). Within this
model, impaired recall of initial and intermediate stimuli is
considered the expression of an inefficient rehearsal process.
Conversely, a reduced recency effect is indicative of a ph-STS
impairment (Vallar & Papagno, 1986).
Phonological similarity effect refers to the tendency of normal subjects to have longer spans for phonologically dissimilar
letters than for phonologically similar letters in both auditory
(Baddeley, 1966; Conrad, 1964; Wickelgren, 1965) and visual
(Silveri et al., 1998) presentations. This effect is linked to the
interference generated by phonological similarity in the ph-STS
(Baddeley & Hitch, 1974) and is not observed when the ph-STS
is defective (Vallar et al., 1997).
Short-term forgetting is the effect of a verbal interpolating
activity on the ability to recall verbal stimuli. The interpolat-

1946

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

Table 4
Word length effect according to input modality
Word length (syllables)

Sequence length
Auditory-repetition

2
4

Visual-repetition

BB

C1

C2

C3

C4

BB

C1

C2

C3

C4

4
5

4
4

4
4

5
4

6
4

6
4

6
5

6
5

5
4

6
5

ing activity is supposed to prevent the articulatory rehearsal
mechanism by minimizing its contribution to recall. Thus, when
subjects are engaged in an interfering activity, their ability to
recall stimuli is considered an index of ph-STS function (Brown,
1958; Peterson & Johnson, 1971; Peterson & Peterson, 1959).
In the auditory-verbal modality, the listened material directly
accesses the ph-STS; thus, the rate of forgetting during an interpolating activity is negligible when the ph-STS is efficient. On
the other hand, when the ph-STS is defective, the rate of forgetting is marked, even after brief interfering activity (Shallice &
Vallar, 1990).
Another index of ph-STS function is the interference of irrelevant speech on immediate verbal recall (unattended speech).
In the auditory modality, both the stimuli to be recalled and
unattended speech have direct access to the ph-STS, with no
contribution of rehearsal (Salamè & Baddeley, 1982). Thus, at
the ph-STS level, unattended speech interferes with the material
to be recalled. When the ph-STS is impaired, irrelevant speech
does not alter immediate verbal recall (Vallar et al., 1997).
6.1. Methods
6.1.1. Participants
For the recency effect, BB’s performance was compared with those of the
Italian normative group described by Capitani, Della Sala, Logie, and Spinnler
(1992). In particular, for the control group we selected subjects with demographic
characteristics similar to those of BB. They included 63 healthy subjects, whose
ages ranged from 50 to 59 years, and whose education level varied from less
than 3 to over 13 years (Capitani et al., 1992).
For the phonological similarity effect and short-term forgetting, four healthy
subjects matched for age (mean age 54 years) and education (18 years) were
tested using the same procedure adopted for the patient.
6.1.2. Materials
6.1.2.1. Recency effect. For free recall, following Silveri et al. (1998), 5 series
of 10 words controlled for length and word frequency were administered.
6.1.2.2. Phonological similarity effect. A modified version of the original test
developed by Brizzolara et al. (2002) was used. The adjusted version consisted of
two series of two-syllable words (Lists 3 and 4 of the original version) arranged in
progressively longer sequences (2–6 items). Stimuli consisted of phonologically
similar and phonologically dissimilar words, which were administered in the
verbal or the visual modality.
6.1.2.3. Short-term forgetting. Brown–Peterson’s paradigm was used (Brown,
1958; Peterson & Peterson, 1959). In this task, 30 three-consonant strings, randomly generated among other phonologically dissimilar letters (F, M, Q, R, S,
H, Z), were presented.
6.1.2.4. Unattended speech. The effect of unattended speech was evaluated in
7-digit sequences. Unattended speech consisted of taped two- and three-letter
nonsense syllables and included a variety of C–V arrangements (CV, CCV, CVC)

presented at one syllable per second, following the protocol of Burani et al.
(1991).
6.1.3. Design and procedure
6.1.3.1. Recency effect. This test consisted of five consecutive immediate freerecall trials. In the first immediate trial, the examiner presented 10 words orally
at 1 every 2 s. Immediately after the presentation, the subject was asked to
recall as many words as possible. Following Silveri et al. (1998), we considered
the recollection of the first 2 items a measure of the primacy effect, and the
recollection of the last 2 items a measure of the recency effect.
6.1.3.2. Phonological similarity effect. These test items were organised in progressively longer series (2–6 items). Each string of words was presented at one
word per second, and immediately after presentation, the subject had to repeat
each series of words in the correct order. If he succeeded, a longer series was
presented. Conversely, if the subject failed on a sequence, a second series of the
same length was administered. The test was discontinued when the subject failed
on two successive same-length strings. The span was calculated as the number
of items in the longest correctly recalled string. In the auditory modality, the
stimuli were read aloud by the examiner; in the visual modality, printed strings
of words were presented on white cards.
6.1.3.3. Short-term forgetting. The examiner presented three-consonant strings
in the auditory modality. The subject had to repeat the consonant sequence
of each string in the same order in which it was administered. The task was
presented under two conditions. In the first condition, the subject had to respond
after varying intervals (5, 10, and 20 s) without engaging in any concurrent task.
In the second condition, the response was given after different intervals (5,
10, and 20 s), which were filled by an interpolated activity (interference). In the
interference condition, the subject was instructed to serially subtract 3 starting
from a number randomly chosen between 100 and 999.
There were five trials for each of the six experimental conditions (three
retention intervals × two concurrent tasks). For each condition, the score was
calculated as the total number of letters correctly recalled (range 0–15). BB’s performance without or with 5 s of interpolating activity was compared by frequency
distribution test (χ2 -value).
6.1.3.4. Unattended speech. The taped unattended speech was started before
auditory presentation of the digit sequences and continued throughout the memory task. In this paradigm, we used the digit sequence length at which the patient
reached at least 70% correct recognition in the standard auditory-pointing span
task (see Table 3). The task was conducted in two sessions. In each session,
the patient received two blocks of stimuli to compare interference (I+) and control conditions (I−) in an ABBA design: interference (I+), control (I−), control
(I−), and interference (I+). The span procedure was auditory-verbal. BB’s performance with or without auditory interference was compared by frequency
distribution test (χ2 -value).

6.2. Results
6.2.1. Recency effect
In the immediate free recall of words, the patient remembered
8 of 10 stimuli in the first 2 positions (primacy effect), 16 of 30
items in the intermediate positions (3–8 positions), and 5 of

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

Fig. 2. BB’s primacy and recency effects in the free recall of words: percentage
of correct stimuli recalled according to their serial positions.

10 stimuli in the last 2 positions (recency effect). Thus, a clear
advantage for the retention of the initial items (primacy effect)
was observed compared with no advantage for the retrieval of the
last stimuli (recency effect) on the list. The noticeable reduction
in the recency effect is consistent with a defect in the ph-STS
(Fig. 2).

1947

Fig. 3. Percentage of recalled auditory-verbal stimuli on Brown–Peterson
paradigm in free condition and in suppression articulatory condition (time delay:
5 s, 10 s, 20 s).

task, the control performance remained substantially unchanged,
while BB’s rate of forgetting increased progressively (Fig. 3).
The effect of the interferential activity was already evident
within 5 s (χ2 = 7.50, d.f. = 1, p = 0.00), indicating quick forgetting of the material within a few seconds after presentation of the
stimuli and confirming the hypothesis of an impaired ph-STS.

6.2.2. Phonological similarity effect
BB did not show a phonological similarity effect in either the
auditory condition or the visual condition, because his spans on
phonologically similar and dissimilar words were comparable.
Conversely, as expected, control subjects tended to recall phonologically dissimilar stimuli better than phonologically similar
ones. It is worth noting that the relationship between ph-STS
impairment and the similarity effect is not linear.
The absence of both auditory and visual phonological similarity effects (Silveri & Cappa, 2003), as well as presence of
auditory but not visual phonological similarity effects (Vallar,
Papagno, & Baddeley, 1991), has been reported in patients
with ph-STS impairment. Based on the interpretations of Vallar
et al. (1997) and Silveri and Cappa (2003), we assumed that
the absence of an evident phonological similarity effect with
auditory-verbal stimuli was indicative of a disorder of the phSTS. Patient and control subject performances are reported in
Table 5.

6.2.4. Unattended speech
BB showed no effect of unattended speech. He recalled 16
of 20 seven-item lists (80%) correctly during auditory interference and 18 (90%) of 20 under the control condition (χ2 = 0.78,
d.f. = 1, p = 0.37).
The results of Experiment 3 confirm the absence of a phonological similarity effect with auditory input, a reduced recency
effect in the auditory free recall of words, no effect of unattended
speech, and a noticeable forgetfulness after 5 s of interference
delay in the Brown–Peterson paradigm. Overall, the patient’s
performance is indicative of a defective ph-STS and a preserved
rehearsal system.
However, the absence of a phonological similarity effect
in the visual modality raises doubts about the function of the
rehearsal system. On the other hand, it could also indicate a
problem at the level of the ph-STS. It has been proposed that
patients with a ph-STS impairment may strategically choose not
to use the rehearsal mechanism, given that rehearsing material
stored in a dysfunctional ph-STS does not significantly improve
their performance (Vallar et al., 1997).

6.2.3. Short-term forgetting
In the free condition, patient performance after each interval
did not differ from the controls (Fig. 3). During the interpolating

7. Experiment 4: lexical–semantic influence on verbal
immediate serial recall

Table 5
Phonological similarity effect according to input modality
Stimuli

Sequence length
Auditory-repetition

Ph-similar
Ph-dissimilar

Visual-repetition

BB

C1

C2

C3

C4

BB

C1

C2

C3

C4

4
4

3
5

2
4

3
4

4
5

4
4

5
6

5
6

4
5

4
5

Results obtained in the previous experiments indicate that
BB’s deficit is confined to the ph-STS, while his use of the
rehearsal process is normal. A defective ph-STS will conceivably affect both word and nonword spans. Nevertheless, BB
demonstrated clearly defective performance with only nonword
stimuli. How can BB maintain good word span performances if
the ph-STS is defective?
An experiment was designed to investigate whether
lexical–semantic coding represents a compensatory mechanism

1948

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

that allows good word span performance. To this aim, word
and nonword-serial recall tasks were administered, with a delay
interval that was either filled or unfilled by an interpolated
activity. Interpolated activity disrupts the subvocal rehearsal
but not the support of lexical–semantic long-term knowledge
to ph-STS performance. Thus, comparison of word and nonword scores during interference tasks will evaluate the influence
of the lexical–semantic component on verbal immediate recall
independently from the function of the rehearsal.
7.1. Methods
7.1.1. Participants
In addition to BB, four healthy subjects matched for age (mean age: 54 years)
and education (18 years) were tested.
7.1.2. Materials
Three different sets of items were used. The first and second set of stimuli
included 6 lists of two-syllable words each (semantically related or unrelated).
Semantically related words belonged to the categories of food, vehicles, animals,
body parts, clothes, and colours. The third set of stimuli consisted of 2 lists of
nonwords comprising five- or six-letter strings.
7.1.3. Design and procedure
After a 20-s interval from the presentation, subjects had to repeat a series of
auditory stimuli in the same order in which they had been administered. Each
set of items was given under two different conditions. In the first condition, the
response was requested after a delay filled by an interference activity (a counting
task). In the second condition, the response was given after an interval unfilled
by the interfering activity.

7.2. Results
When the task did not require interference activity and BB
could normally refresh the stimuli, his performance was preserved using every list, independent of the presence of the
semantic component. In fact, in the semantically related list, BB
recalled 34 of 36 words (94%) compared with 34, 32, 32 and 31
stimuli (94%, 86%, 86%, and 89%, respectively) for the control
subjects, and in the semantically unrelated list, he retrieved 27 of
36 words (75%) compared with 24, 24, 24 and 28 stimuli (67%,
67%, 67% and 78%, respectively) for the controls.
With regard to nonwords, BB recalled 7 of 8 items (87%), and
control subjects recalled 6, 7, 7 and 7 of 8 items (75%, 87%, 87%
and 87%, respectively). Conversely, when BB could not refresh
the stimuli because of the interference activity, his performance
was poor.
Overall, BB recalled fewer items than control subjects and
showed a striking difference in nonword stimuli. Indeed, with
interference activity BB recalled 26 of 36 semantically related
words (72%) in comparison with 27, 32, 32, and 33 words (75%,
89%, 89%, and 92% respectively) in control subjects; 19 of 36

unrelated stimuli (53%) compared with 20, 21, 22 and 29 items
(55%, 58%, 61% and 78%, respectively) in the normal subjects;
and 1 of 8 nonwords (12%) compared with 4, 4, 5 and 5 stimuli
(50%, 50%, 62% and 62%, respectively) for the controls.
The differences between BB’s performance with word and
nonword stimuli were further highlighted by frequency distribution test (χ2 -value). BB recalled a significantly lower
number of semantically related and unrelated words (χ2 = 6.40,
d.f. = 1, p = 0.01 and χ2 = 3.85, d.f. = 1, p = 0.04) and nonwords
(χ2 = 6.25, d.f. = 1, p = 0.01) when he was engaged in an interference activity than when he was not. Furthermore, taking
into account both tasks performed with interference activity,
BB’s performance was significantly better with word (semantically unrelated or related) than with nonword stimuli (χ2 = 4.28,
d.f. = 1, p = 0.03; χ2 = 9.85; d.f. = 1, p = 0.00), indicating an ameliorative effect of the semantic component. Patient and control
scores are reported in Table 6.
The efficacy of the interpolating activity in reducing BB’s
word and nonword performances demonstrates the high efficiency of his rehearsal system and confirms that the ph-STS is
the pathological locus of his vWM impairment. Furthermore, the
difference between word and nonword recollection in the presence of interpolating activity demonstrates that lexical–semantic
knowledge is able to support the phonological trace, even in an
impaired ph-STS.
Different theories support the idea that long-term knowledge plays a role in the immediate recall of words. The
interactive activation models (Martin, Lesch, & Bartha, 1999;
Martin & Saffran, 1997) suggest that verbal short-term memory corresponds to the temporary activation of semantic and
phonological representations in the language production system.
“During immediate serial recall tasks all linguistic representations (including lexical and semantic representations) would be
activated via bi-directional connections between levels of representations which in turn support recall performance” (Caza,
Belleville, & Gilbert, 2002). Thus, when a word is presented
verbally, its phonological representation is activated first, but
activation then spreads to the corresponding semantic and lexical
nodes (Forde & Humphreys, 2002).
In contrast to the view that long-term semantic representations normally form part of verbal short-term memory, “the
redintegration theory” suggests that semantic memory only
plays a role when information in short-term memory has been
disrupted (Hulme et al., 1997). According to this theory, when
a phonological trace is partially degraded, representations permanently stored in long-term memory can be recalled in order
to “redintegrate” or “clean up” the trace in short-term memory (Poirier & Saint-Aubin, 1995; Schweickert, 1993; Walker
& Hulme, 1999). This mechanism may explain why traces of

Table 6
Percentage of recalled stimuli in word (semantically related/unrelated) and nonword repetition with and without interference task
Conditions

Interference
No interference

Nonwords

Related words

Unrelated words

BB

C1

C2

C3

C4

BB

C1

C2

C3

C4

BB

C1

C2

C3

C4

12
87

50
75

62
87

50
75

62
71

72
94

92
94

89
86

89
86

75
89

53
75

55
67

78
78

58
67

61
67

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

words that have meaning-based features are more resistant to
forgetting (Plaut & Shallice, 1993).
8. General discussion
The present findings demonstrate the presence of a vWM
impairment in a patient with a lesion confined to the cerebellum
due to a selective defect of the ph-STS with preservation of the
rehearsal system function.
Directly after the cerebellar ischemic lesion, BB performed
pathologically on immediate and delayed verbal serial recall and
auditory recognition tasks. Because no lesions were detected
in the supratentorial structures and because the only lesioned
structure was the cerebellum, we can reasonably assume that
the vWM impairment was related to the cerebellar damage.
BB presented with a vWM impairment with a normal verbal
span. Justus et al. (2005) noted a largely preserved digit span
in patients with degenerative or focal cerebellar damage, confirming previous observations that the digit span of degenerative
cerebellar subjects was comparable with that of controls (Burk
et al., 2003; Fabbro et al., 2004; Globas et al., 2003; Le Pira et
al., 2002). Even when lower span scores were reported for cerebellar patients compared with controls, they were still within the
normal range (Ravizza et al., 2006).
As expected with cerebellar lesions, BB’s verbal recall
deficits showed good recovery (Dunwoody, Alsagoff, & Yuan,
1997; Pollack, 1997; Silveri, Leggio, & Molinari, 1994; Silveri
et al., 1998). At the time of our study, 17.6 months after the
lesion, his performance on verbal serial recall tasks was within
normal limits. However, even though span values were still normal for digit and word stimuli, a clear defect was present for
nonword stimuli.
Because pathological performance on nonword stimuli can
reflect vWM malfunction, we investigated BB’s vWM system
further by analysing each component of the phonological loop.
On digit span tasks (Experiment 1), BB’s overall performance
was normal. However, an advantage was present in the visualpointing modality with respect to the other conditions, indicating
good phonological recoding and questioning phonological output buffer functioning.
We found that the rehearsal process recirculated verbal information effectively (Experiment 2), as demonstrated by the strong
effect of articulatory suppression and by the presence of a
word length effect with visual stimuli. However, the lack of a
word length effect with auditory stimuli raises doubts about the
integrity of the rehearsal system.
Nevertheless, there is evidence supporting good rehearsal
function in spite of the absence of a word length effect. Firstly,
two of our control subjects did not present the auditory word
length effect. Moreover, there are reports of patients with no
word length effect who presented a selective deficit of the phSTS with a preserved rehearsal process (see patient LA, in Silveri
& Cappa, 2003; Vallar et al., 1997). This phenomenon has been
interpreted as a strategic choice in the presence of a dysfunctional ph-STS, considering that rehearsing the material does
not substantially improve memory performance (Vallar et al.,
1997).

1949

BB’s pattern of ph-STS function (Experiment 3) was consistent with selective ph-STS impairment. The reduced recency
effect in immediate free recall and the absence of a phonological similarity effect with auditory stimuli suggest a defect of
the ph-STS. The insensitivity of BB’s span to the interfering
effect of unattended speech also indicates a ph-STS impairment
(Vallar et al., 1997). In the Brown–Peterson paradigm, the interpolating activity prevents the operation of rehearsal (Brown,
1958; Peterson & Johnson, 1971; Peterson & Peterson, 1959).
Consistent with the hypothesis of a ph-STS defect, in the free
condition the patient was able to retain almost all auditory stimuli; however, his performance was poor even after only 5 s of
interpolating activity.
In summary, BB’s scores decreased noticeably on tasks that
primarily or exclusively engaged activity of the ph-STS, while
his performance was completely preserved on tests that investigate the recirculation of verbal information. This pattern of
results implicates the ph-STS as the cognitive locus of the
patient’s deficit.
A possible limitation of the work, common to all single case
studies, is the difficulty of providing statistically relevant data. In
the present study, as in many other single case studies, caution is
required when interpreting results, which will need confirmation
from group studies.
BB’s performance on verbal span tasks was normal for digit
and word stimuli and markedly defective for nonword stimuli.
This evidence supports the hypothesis that differential function
of the ph-STS depends on characteristics of the material processed. In Experiment 4, word- and nonword-serial recall tasks
with interpolating activity were performed. In this condition,
BB’s performance was significantly improved by the presence
of meaningful items.
As already described, when the rehearsal mechanism is
blocked by interpolating activities, the recall of auditoryverbal stimuli is considered a measure of ph-STS integrity
(Graceffa, Carlesimo, Peppe, & Caltagirone, 1999). Because
defective scores on nonword-serial recall tasks were obtained
only when the rehearsal mechanism was blocked, this finding
is surely linked to ph-STS impairment. The obvious difference
between word and nonword stimuli is the semantic component.
Therefore, it is conceivable to assume that lexical–semantic
knowledge facilitates the retrieval of phonological information.
In the case of BB, the data indicate that deficits are evident only
if all of the ancillary mechanisms that support the ph-STS – i.e.,
rehearsal mechanisms and lexical–semantic compensation – are
blocked.
Another aspect that deserves discussion is the dissociation
between impaired performance in delayed auditory-verbal cued
recognition and normal performance in visual cued recognition
of the TOMAL. Several studies have demonstrated that concrete
words are highly imaginable and promptly recalled from longterm memory (Davelaar & Besner, 1988; Paivio, 1966; Paivio,
Yuille, & Smythe, 1966). According to the ‘dual-coding’ theory
(Paivio, 1986), imaginable words can be stored by using both
verbal and visual codes (Binder et al., 1997; Demb et al., 1995;
Petersen, Fox, Posner, Mintun, & Raichle, 1988; Vanderberghe,
Price, Wise, Josephs, & Frackowiak, 1996). Thus, concrete

1950

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

words may rely on multiple forms of representation that facilitate
their recollection (Baddeley, 2000).
Based on BB’s performance, we hypothesise that the patient
performed visual cued delayed recognition correctly by activating semantic components of the prefrontal cortex through
the generation of perceptual and mental images. Thus, even
if ph-STS impairment hindered BB’s retrieval of phonological
information, visual cues allowed him to generate a percept and
a mental image of the concrete word. This mechanism cannot
be active with auditory cues that require ph-STS activity; thus,
delayed auditory-verbal cued recognition was impaired.
Functional neuroimaging data have consistently reported
cerebellar activation in vWM tasks, but with differences in the
lobular distribution of the activation foci. In 1996, Fiez and
colleagues (Fiez et al., 1996) reported bilateral cerebellar activation in relation to short-term maintenance of verbal information;
however, a subsequent report (Chein & Fiez, 2001) linked this to
the first stage of the encoding phase, suggesting that it may not
be engaged during rehearsal per se. Similarly, bilateral anterior
and posterior cerebellar activation was described during digit
span tasks under conditions of reduced rehearsal (Gerton et al.,
2004).
Desmond et al. reported several foci of cerebellar activation
in verbal working memory tasks (Desmond et al., 1997)—i.e.,
bilaterally in the superior cerebellar hemispheres, in the posterior vermis, and in the right inferior cerebellar hemisphere.
On the basis of the cerebro-cerebellar connections, the authors
connected activation of the right inferior cerebellar cortex to
phonological store activity and activation of the superior cerebellar cortex to the articulatory control process.
Event-related fRMI data indicate that these two cerebellar
regions are functionally different (Chen & Desmond, 2005a,
2005b). The right superior cerebellum (lobules VI/Crus1) was
strongly activated during the initial encoding phase of a verbal working memory task, while the right inferior cerebellum
(lobules VIIB/VIIIA) showed the greatest activation during the
maintenance phase (Chen & Desmond, 2005a, 2005b).
Discrepancies in neuroimaging data may arise from the existence of a distributed cerebellar network whose activation varies
in relation to mnesic load (Desmond et al., 1997; Zhang et al.,
2003), input modality (Schumacher et al., 1996), or category
(Ravizza et al., 2006), and particularly according to different
phases of the working memory process (Chen & Desmond,
2005a, 2005b). Although neuroimaging data overall demonstrate the importance of the cerebellum in vWM, they do not
allow clarification of the specific role of the different cerebellar
folia in vWM functions (Justus et al., 2005).
BB presented two different foci of cerebellar damage—one
localized in the left postero-inferior cerebellum (lobule VIIIA)
and one in the right superior cerebellum (lobule V). Lobule
VIIIA is within the area defined by several studies as involved in
vWM (Chen & Desmond, 2005a, 2005b; Desmond et al., 1997).
Although Chen and Desmond hypothesised right functional lateralization, this model has been contradicted by different reports
that have indicated bilateral activation (Chein & Fiez, 2001; Fiez
et al., 1996; Gerton et al., 2004). The importance of the inferior
cerebellar lobe (without clear lateralization) in vWM is also sup-

ported by a recent study that did not find a correlation between
the laterality of the damaged inferior cerebellar lobe and digit
span performance (Ravizza et al., 2006).
The functional contribution of cerebellar processing to vWM
requires further consideration. On the basis of neuroimaging (Paulesu et al., 1993; Petrides et al., 1993), as well as
neurophysiological (Desmond, Chen, & Shieh, 2005) and neuropsychological studies (Silveri et al., 1998), the cerebellar role
in vWM was first connected to the rehearsal process. However, the articulatory rehearsal hypothesis has been criticised
on the basis of neuroimaging (Awh et al., 1996; Chein & Fiez,
2001) and neuropsychological observations (Justus et al., 2005;
Ravizza et al., 2006), and accordingly, alternative hypotheses
have been advanced.
Desmond and colleagues (Desmond, 2001; Desmond et
al., 1997) proposed that the cerebellum serves as an interface between the ph-STS and articulatory rehearsal, comparing
the output of subvocal articulation with the contents of the
phonological store. Because smooth and rapid updating of the
phonological store requires predictive control of the articulatory process, “the cerebellum could compute the discrepancy
between actual and intended phonological rehearsal and use this
information to update a feedforward command to the frontal
lobe, thereby facilitating the phonological loop” (Desmond et
al., 1997).
In a report on the motor theory of speech, which emphasizes the relationships between perception and action (Liberman
& Mattingly, 1985), Ravizza et al. (2006) argued that cerebellar processing in vWM may be important for integrating
acoustic/phonetic representations with articulatory representations and for correcting degraded sensory information. To this
end, the cerebellum may provide a mechanism for creating more
integrated and stronger memory traces, which are less prone to
decay and error.
Ravizza et al. (2006) hypothesised a role of the cerebellum
in vWM that was not limited to merely the rehearsal process
but was also involved with the ph-STS. The present report is
the first to demonstrate a selective impairment of the ph-STS,
with sparing of the rehearsal process after cerebellar damage.
Considering previous lesion studies that have reported selective
impairment of the rehearsal component of the vWM, it is conceivable to hypothesise that also in vWM, cerebro-cerebellar
interactions may act according to parallel segregated loops, as
hypothesised for motor and cognitive functions (Middleton &
Strick, 1997).
Further, as suggested also by Chen and Desmond (2005a),
different cerebro-cerebellar loops support the functions of different cerebral modules. Therefore, isolated cerebellar lesions
might affect one modular function with only discrete effects on
performance as a whole.
How does this mechanism apply to the Baddeley’s model of
vWM? Each of the vWM cortical modules may receive in support selective cerebellar channels. If this is the case, differences
in anatomical localisation of cerebellar damage can affect one
channel while sparing the others. Thus, cerebellar patients may
present different modalities of vWM impairment according to
the topography of their lesion.

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

In this context, the discrepancy between the neuroimaging
literature, which has shown consistent involvement of the cerebellum in vWM, and the neuropsychological literature, which
has reported very little evidence of clear vMW impairment in
patients with cerebellar damage, is readily apparent. In fact, in
working memory function, as in other cognitive domains, cerebellar damage can be compensated by other neural systems that
are still able to perform the function, but at a less efficient level.
Is the cerebellar contribution limited to verbal working memory or does it intervene also in spatial working memory? This
question is interesting, and of relevance for general theories on
the cerebellar functions. Present data do not allow us to directly
address this topic. We observed normal verbal and spatial working memory performances in the general neuropsychological
examination (see Tables 1 and 2). Nevertheless, when vWM has
been addressed in detail, clear impairments have been demonstrated. Thus, it may be argued that spatial working memory
impairments might also be revealed by a more detailed analysis. At present only few studies addressed the cerebellar role
in spatial working memory with contrasting data. Studies in
rodents reported various degrees of cerebellar involvement in
spatial working memory (Mandolesi, Leggio, Graziano, Neri, &
Petrosini, 2001; Martin, Escher, Goldowitz, & Mittleman, 2004)
while excitotoxic lesions of deep cerebellar nuclei in monkeys
did not affect spatial working memory (Nixon & Passingham,
1999). In humans, early functional imaging studies did not find
cerebellar activation in spatial working memory tasks (Smith &
Jonides, 1998) and the cerebellum has been suggested not to play
a role in this function (Ivry & Fiez, 2000). Nevertheless, more
recent functional imaging data indicated cerebellar activations
associated with spatial working memory tasks (Haberecht et al.,
2001; Koppelstaetter et al., 2008; Luna et al., 2002). Thus, cerebellum and spatial working memory appear promising fields for
future research.
Acknowledgements
We thank BB for the hours of testing he patiently endured.
The continuous encouragement and support of Professor Carlo
Caltagirone and the professional English style editing of Claire
Montagna are gratefully acknowledged. The editing support of
Blue Pencil Science LLC is also acknowledged. The present
work was supported in part by grants to MM and MGL from
MIUR and the Italian Ministry of Health.
References
Andreasen, N., O’Leary, D., Arndt, S., Cizadlo, T., Hurtig, R., Rezai, K., et
al. (1995). Short-term and long-term verbal memory: A positron emission
tomography study. Proceedings of the National Academy of Sciences of the
United States of America, 92, 5111–5115.
Awh, E., Jonides, J., Smith, E., Schumacher, E., Koeppe, R., & Katz, S. (1996).
Dissociation of storage and rehearsal in verbal working memory: Evidence
from position emission tomography. Psychological Science, 7, 25–31.
Baddeley, A. (1966). Short-term memory for word sequences as a function of
acoustic, semantic and formal similarity. Quarterly Journal of Experimental
Psychology, 18, 362–365.
Baddeley, A. (1990). Human memory: Theory and practice. Hover: Erlbaum.

1951

Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4, 417–423.
Baddeley, A. (2003). Working memory: Looking back and looking forward.
Nature Review of Neuroscience, 4, 829–839.
Baddeley, A., & Hitch, G. (1974). Working memory. In G. H. Bower (Ed.), The
psychology of learning and motivation. Advances in research and theory
(pp. 47–89). New York: Academic Press.
Baddeley, A., Lewis, V., & Vallar, G. (1984). Exploring the articulatory loop.
Quarterly Journal of Experimental Psychology, 36A, 233–252.
Baddeley, A., Thomson, N., & Buchanan, M. (1975). Word length and the structure of short-term memory. Journal of Verbal Learning and Verbal Behavior,
14, 575–589.
Binder, J., Frost, J., Hammeke, T., Cox, R., Rao, S., & Prieto, T. (1997). Human
brain language areas identified by functional magnetic resonance imaging.
Journal of Neuroscience, 17, 362.
Borkowsky, J., Benton, A., & Spreen, O. (1967). Word fluency and braindamage. Neuropsychologia, 5, 135–140.
Brizzolara, D., Casalini, C., Chilosi, A., & Cipriani, A. (2002). La memoria
di lavoro nei disturbi di apprendimento. In S. Vicari & M. Caselli (Eds.), I
Disturbi dello sviluppo. Neuropsicologia clinica e ipotesi riabilitative (pp.
241–260). Bologna: Il Mulino.
Brown, J. (1958). Some tests of the decay theory of immediate memory. Journal
of Experimental Psychology, 10, 12–21.
Brown, G., & Hulme, C. (1995). Modelling item length effects in memory span:
No rehearsal needed? Memory and Language, 34, 594–621.
Burani, C., Vallar, G., & Bottini, G. (1991). Articulatory coding and phonological judgements on written words and pictures: The role of the phonological
output buffer. The European Journal of Cognitive Psychology, 3, 379–
398.
Burk, K., Globas, C., Bosch, S., Klockgether, T., Zuhlke, C., Daum, I., et al.
(2003). Cognitive deficits in spinocerebellar ataxia type 1, 2 and 3. Journal
of Neurology, 250, 207–211.
Caltagirone, C., Gainotti, G., Carlesimo, G., Parnetti, L., Fadda, L., Gallassi, R.,
et al. (1995). The Mental deterioration battery: Description of an instrument
of the neuropsychological diagnosis. Archivio di Psicologia, Neurologia e
Psichiatria, 56, 461–470.
Capitani, E., Della Sala, S., Logie, R., & Spinnler, H. (1992). Recency, primacy
and memory: Rappraising and standardising the serial position curve. Cortex,
28, 315–342.
Caza, N., Belleville, S., & Gilbert, B. (2002). How loss of meaning with preservation of phonological word form affects immediate serial recall performance:
A linguistic account. Neurocase, 8, 255–273.
Chein, J., & Fiez, J. (2001). Dissociation of verbal working memory system
components using a delayed serial recall task. Cerebral Cortex, 11, 1003–
1014.
Chen, S., & Desmond, J. (2005a). Cerebrocerebellar networks during articulatory rhearsal and verbal rehearsa and verbal working memory tasks.
Neuroimage, 24, 332–338.
Chen, S., & Desmond, J. (2005b). Temporal dynamics of cerebro-cerebellar network recruitment during a cognitive task. Neuropsychologia, 43, 1227–1237.
Conrad, R. (1964). Acoustic confusion in immediate memory. British Journal
of Psychology, 55, 75–84.
Davelaar, E., & Besner, D. (1988). Word identification: Imageability, semantics and the content-functor distinction. Quarterly Journal of Experimental
Psychology A, 40, 789–799.
Demb, J., Desmond, J., Wagner, A., Vaidya, C., Glover, G., & Gabrieli, J. (1995).
Semantic encoding and retrieval in the left inferior prefrontal cortex: A
functional MRI study of task difficulty and process specificity. Journal of
Neuroscience, 15, 5870–5878.
Desmond, J. (2001). Cerebellar involvement in cognitive function: Evidence
from neuroimaging. International Review of Psychiatry, 13, 283–294.
Desmond, J. E., Chen, A. H., & Shieh, P. B. (2005). Cerebellar transcranial
magnetic stimulation impairs verbal working memory. Annals of Neurology,
58, 553–560.
Desmond, J., Gabrieli, J., Wagner, A., Ginier, B., & Glover, G. (1997). Lobular
patterns of cerebellar activation in verbal working-memory and fingertapping tasks as revealed by functional MRI. The Journal of Neuroscience,
17, 9675–9685.

1952

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953

Dunwoody, G., Alsagoff, Z., & Yuan, S. (1997). Cerebellar mutism with subsequent dysarthria in an adult: Case report. British Journal of Neurosurgery,
11, 161–163.
Fabbro, F., Tavano, A., Corti, S., Bresolin, N., De Fabritiis, P., & Borgatti, R.
(2004). Long-term neuropsychological deficits after cerebellar infarctions
in two young adult twins. Neuropsychologia, 42, 536–545.
Fiez, J., Raife, E., Balota, D., Schwarz, J., Raichle, M., & Petersen, S. (1996). A
positron emission tomography study of the short-term maintenance of verbal
information. Neuroscience, 16, 808–822.
Forde, E., & Humphreys, G. (2002). The role of semantic knowledge in shortterm memory. Neurocase, 8, 13–27.
Gerton, B., Brown, T., Meyer-Lindenber, A., Kohn, P., Holt, J., Olsen, R., et
al. (2004). Shared and distinct neurophysiological components of the digits forward and backward tasks as revealed by functional neuroimaging.
Neuropsychologia, 42, 1781–1787.
Globas, C., Bosch, S., Zuhlke, C., Daum, I., Dichgans, J., & Burk, K. (2003). The
cerebellum cognition. Intellectual function in spinocerebellar ataxia type 6
(SCA 6). Journal of Neurology, 250, 1482–1487.
Goerlich, C., Daum, I., Hertrich, I., & Ackermann, H. (1995). Verbal shortterm memory and motor speech processes in Broca’s aphasia. Behavioural
Neurology, 8, 81–95.
Graceffa, A., Carlesimo, G., Peppe, A., & Caltagirone, C. (1999). Verbal working
memory deficit in Parkinson’s disease subjects. European Neurology, 42,
90–94.
Grasby, P., Frith, C., Friston, K., Simpson, J., Fletcher, P., Frackowiak, R., et
al. (1994). A graded task approach to the functional mapping of brain areas
implicated in auditory-verbal memory. Brain, 117, 1271–1282.
Haberecht, M. F., Menon, V., Warsofsky, I. S., White, C. D., Dyer-Friedman,
J., Glover, G. H., et al. (2001). Functional neuroanatomy of visuo-spatial
working memory in Turner syndrome. Human Brain Mapping, 14, 96–107.
Hulme, C., Roodenrys, S., Schweickert, R., Brown, G., Martin, M., & Stuart, G.
(1997). Word-frequency effects on short-term memory tasks: Evidence for
a redintegration process in immediate serial recall. Journal of Experimental
Psychology: Learning Memory and Cognition, 23, 1217–1232.
Ivry, R. B., & Fiez, J. A. (2000). Cerebellar contributions to cognition and
imagery. In M. S. Cazzaniga (Ed.), The new cognitive neuroscience (2nd
ed., pp. 999–1011). Cambridge, MA: MIT Press.
Justus, T., Ravizza, S., Fiez, J., & Ivry, R. (2005). Reduced phonological similarity effects in patients with damage to the cerebellum. Brain and Language,
95, 304–318.
Kirschen, M., Chen, S., & Schraedley-Desmond, P. (2005). Load and practicedependent increases in cerebro-cerebellar activation in verbal working
memory: An fMRI study. Neuroimage, 24, 462–472.
Koppelstaetter, F., Poeppel, T. D., Siedentopf, C. M., Ischebeck, A., Verius, M.,
Haala, I., et al. (2008). Does caffeine modulate verbal working memory
processes An fMRI study. Neuroimage, 39, 492–499.
Le Pira, F., Zappala, G., Saponara, R., Domina, E., Restivo, D., Reggio, E., et
al. (2002). Cognitive findings in spinocerebellar ataxia type 2: Relationship
to genetic and clinical variables. Journal of the Neurological Sciences, 201,
53–57.
Levy, B. (1971). Role of articulation in auditory and visual short-term memory.
Journal of Verbal Learning and Verbal Behavior, 10, 123–132.
Li, Y., Chiaravalloti, N., Hillary, F., DeLuca, J., Liu, W., Kalnin, A., et al. (2004).
Differential cerebellar activation on functional magnetic resonance imaging
during working memory performance in persons with multiple sclerosis.
Archives of the Physical Medicine and Rehabilitation, 85, 635–639.
Liberman, A., & Mattingly, I. (1985). The motor theory of speech perception
revised. Cognition, 21, 1–36.
Luna, B., Minshew, N. J., Garver, K. E., Lazar, N. A., Thulborn, K. R., Eddy,
W. F., et al. (2002). Neocortical system abnormalities in autism: An fMRI
study of spatial working memory. Neurology, 59, 834–840.
Maddox, W., Aparicio, P., Marchant, N., & Ivry, R. (2005). Rule-based category
learning is impaired in patients with Parkinson’s disease but not in patients
with cerebellar lesions. Journal of Cognitive Neuroscience, 17, 707–723.
Mandolesi, L., Leggio, M. G., Graziano, A., Neri, P., & Petrosini, L. (2001).
Cerebellar contribution to spatial event processing: Involvement in procedural and working memory components. European Journal of Neuroscience,
14, 2011–2022.

Martin, L. A., Escher, T., Goldowitz, D., & Mittleman, G. (2004). A relationship
between cerebellar Purkinje cells and spatial working memory demonstrated in a lurcher/chimera mouse model system. Genes Brain Behaviour,
3, 158–166.
Martin, R. C., Lesch, M. F., & Bartha, M. C. (1999). Independence of input and
output phonology in word processing and short-term memory. Journal of
Memory and Language, 41, 3–29.
Martin, N., & Saffran, E. M. (1997). Language and auditory-verbal short-term
memory impairments: Evidence for common underlying processes. Cognitive Neuropsychology, 14, 641–682.
Miceli, G., Laudanna, A., & Burani, C. (1994). Batteria per l’analisi dei deficit
afasici- BADA-. Roma: CEPSAG.
Middleton, F., & Strick, P. (1997). Dentate output channels: Motor and cognitive
components. Progress in Brain Research, 114, 553–566.
Murray, D. (1968). Articulation and acoustic confusability in short-term memory. Journal of Experimental Psychology, 78, 679–684.
Nixon, P. D., & Passingham, R. E. (1999). The cerebellum and cognition: Cerebellar lesions do not impair spatial working memory or visual associative
learning in monkeys. European Journal of Neuroscience, 11, 4070–4080.
Oldfield, R. (1971). The assessment and the analysis of handedness: the Edinburgh Inventory. Neuropsychologia, 9, 97–113.
Orsini, A., Grossi, D., Capitani, E., Laiacona, M., Papagno, C., & Vallar, G.
(1987). Verbal and spatial immediate memory span: Normative data from
1355 adults and 112 children. Italian Journal of Neurological Science, 8,
539–548.
Paivio, A. (1966). Latency of verbal associations and imagery to noun stimuli as
a function of abstractness and generality. Canadian Journal of Psychology,
20, 378–387.
Paivio, A. (1986). Mental representations: A dual coding approach. New York:
Oxford Academic Press.
Paivio, A., Yuille, J., & Smythe, P. (1966). Stimulus and response abstractness,
imagery and meaningfulness and reported mediators in paired-associate
learning. Canadian Journal of Psychology, 20, 362–377.
Paulesu, E., Frith, C., & Frackowiak, R. (1993). The neural correlates of the
verbal component of working memory. Nature, 363, 583–584.
Petersen, S., Fox, P., Posner, M., Mintun, M., & Raichle, M. (1988). Positron
emission tomographic studies of the cortical anatomy of single-word processing. Nature, 331, 585–589.
Peterson, L., & Johnson, S. (1971). Some effects of minimizing articulation on
short-term retention. Journal of Verbal Learning and Verbal Behavior, 10,
346–354.
Peterson, L., & Peterson, M. (1959). Short-term retention of individual verbal
items. Journal of Experimental Psychology, 58, 193–198.
Petrides, M., Alivisatos, B., Meyer, E., & Evans, A. (1993). Functional activation
of the human frontal cortex during the performance of verbal working memory tasks. Proceedings of the National Academy of Sciences of the United
States of America, 90, 878–882.
Plaut, D., & Shallice, T. (1993). Deep dyslexia: A case study of connectionist
neuropsychology. Cognitive Neuropsychology, 10, 377–500.
Poirier, M., & Saint-Aubin, J. (1995). Memory for related and unrelated words:
Further evidence on the influence of semantic factors in immediate serial
recall. Quarterly Journal of Experimental Psychology A, 48, 384–404.
Pollack, I. (1997). Posterior fossa syndrome. In J. D. Schmahmann (Ed.), The
cerebellum and cognition (pp. 411–432). San Diego: Academic Press.
Raven, J. (1947). Progressive matrices. Set A, Ab, B: Board and book form.
London: Lewis.
Ravizza, S., & Ivry, R. (2001). Comparison of the basal ganglia and cerebellum
in shifting attention. Journal of Cognitive Neuroscience, 13, 285–297.
Ravizza, S., McCormick, C., Schlerf, J., Ivry, R., & Fiez, J. (2006). Cerebellar
damage produces selective deficits in verbal working memory. Brain, 129,
306–320.
Rey, A. (1968). Reattivo della Figura Complessa- Manuale. Firenze, Italy:
Organizzazioni Speciali.
Reynolds, C., & Bigler, E. (1995). Test di Memoria e Apprendimento- Manuale.
Trento: Erickson.
Salamè, P., & Baddeley, A. (1982). Distruption of short-term memory by unattended speech: Implications for the structure of working memory. Journal
of Verbal Learning and Verbal Behavior, 21, 150–164.

F.R. Chiricozzi et al. / Neuropsychologia 46 (2008) 1940–1953
Schmahmann, J., Doyon, J., Toga, A., & Petrides, M. (2000). MRI atlas of the
human cerebellum. San Diego CA: Academic Press.
Schumacher, E., Lauber, E., Awh, E., Jonides, J., Smith, E., & Koeppe, R. (1996).
PET evidence for an amodal verbal working memory system. Neuroimage,
3, 79–88.
Schweickert, R. (1993). A multinomial processing tree model for degradation
and redintegration in immediate recall. Memory and Cognition, 21, 168–175.
Schweickert, R., Guentert, L., & Hersberger, L. (1990). Phonological similarity,
pronunciation rate and memory span. Psychological Science, 1, 74–77.
Shallice, T., & Vallar, G. (1990). The impairment of auditory-verbal short-term
storage. In G. Vallar & T. Shallice (Eds.), Neuropsychological impairments
of short-term memory (pp. 11–53). New York: Cambridge University Press.
Silveri, M., & Cappa, A. (2003). Segregation of the neural correlates of language
and phonological short-term memory. Cortex, 39, 913–925.
Silveri, M., Di Betta, A., Filippini, V., Leggio, M., & Molinari, M. (1998).
Verbal short-term store-rehearsal system and the cerebellum. Evidence from
a patient with a right cerebellar lesion. Brain, 121, 2175–2187.
Silveri, M., Leggio, M., & Molinari, M. (1994). The cerebellum contributes
to linguistic production: A case of agrammatic speech following a right
cerebellar lesion. Neurology, 44, 2407-2050.
Smith, E. E., & Jonides, J. (1998). Neuroimaging analyses of human working memory. Proceedings of the National Academy of Sciences, 95,
12061–12068.
Sperling, G. (1963). A model for visual memory tasks. Human Factors, 5, 19–31.
Stigler, J., Lee, S., & Stevenson, H. (1986). Digit memory in Chinese and
English: Evidence for temporally limited store. Cognition, 23, 1–20.
Vallar, G., Basso, A., & Bottini, G. (1990). Phonological processing and sentence comprehension: A neuropsychological case study. In G. Vallar & T.
Shallice (Eds.), Neuropsychological impairments of short-term memory (pp.
448–476). New York: Cambridge University Press.
Vallar, G., & Cappa, S. (1987). Articulation and verbal short-term memory.
Evidence from anarthria. Cognitive Neuropsychology, 4, 55–78.
Vallar, G., Corno, M., & Basso, A. (1992). Auditory and visual verbal short-term
memory in aphasia. Cortex, 28, 383–389.

1953

Vallar, G., Di Betta, A., & Silveri, M. (1997). The phonological short-term
store-rehearsal system: Patterns of impairment and neural correlates. Neuropsychologia, 36, 795–812.
Vallar, G., & Papagno, C. (1986). Phonological short-term store and the nature
of the recency effect. Evidence from neuropsychology. Brain and Cognition,
5, 428–442.
Vallar, G., & Papagno, C. (1995). Neuropsychological impairments of short-term
memory. In A. D. Baddeley, B. A. Wilson, & F. Watts (Eds.), Handbook of
memory disorders (pp. 135–165). Chichester: John Wiley and Sons.
Vallar, G., Papagno, C., & Baddeley, A. (1991). Long-term recency effects and
phonological short-term memory. A neuropsychological case study. Cortex,
27, 323–326.
Vanderberghe, R., Price, C., Wise, R., Josephs, O., & Frackowiak, R. (1996).
Functional anatomy of a common semantic system for words and pictures.
Nature, 383, 254–256.
Walker, I., & Hulme, C. (1999). Concrete words are easier to recall than abstract
words: Evidence for a semantic contribution to short-term serial recall.
Journal of Experimental Psychology, 25, 1256–1271.
Wechsler, D. (1997). Wais-r Wechsler Intelligence Scale Revised. Traduzione e
adattamento italiano- Manuale. Firenze: Organizzazioni Speciali.
Wickelgren, W. (1965). Short-term memory fot phonemically similar lists. American Journal of Psychology, 78, 567–574.
Witt, K., Nuhsman, A., & Deuschl, G. (2002). Intact artificial grammar learning
in patients with cerebellar degeneration and advanced Parkinson’s disease.
Neuropsychologia, 40, 1534–1540.
Zazzo, R. (1980). Le test des deux barrages. Firenze: Organizzazioni Speciali.
Zhang, D., Li, Z., Chen, X., Wang, Z., Zhang, X., Meng, X., et al. (2003).
Functional comparison of primacy, middle and recency retrieval in human
auditory short-term memory: An event-related fMRI study. Cognitive Brain
Research, 16, 91–98.
Zhang, G., & Simon, H. (1985). STM capacity for Chinese words and idioms:
chunking and acoustical loop hypotheses. Memory and Cognition, 13,
193–201.