BJMB
Brazilian Journal of Motor Behavior
Special issue:
The role of practice in motor learning
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
153 of 161
Effects of the different distributed practice regimes on the learning of three-ball cascade
juggling task
JOSÉ E. M. LUZ
1
| HENRIQUE D. SANTOS
1
| GIORDANO M. G. BONUZZI
1,2
1
Departament of Physical Education, State University of Piauí, Picos, Piaui, Brazil.
2
Departament of Physical Education, Federal University of Vale do São Francisco, Petrolina, Pernambuco, Brazil.
Correspondence to: Giordano Marcio Gatinho Bonuzzi. Professor Barros Araújo campus, BR-316, KM 299, Altamira, Picos, Piaui, Brazil. Zipcode: 64602-000
email: [email protected]gmail.com
https://doi.org/10.20338/bjmb.v16i2.268
ABBREVIATIONS
AMONG Among days distributed practice
group
MASSED Massed practice group
WITHIN Within-one-day distributed practice
group
PUBLICATION DATA
Received 13 11 2021
Accepted 30 01 2022
Published 01 06 2022
BACKGROUND: In its majority, the literature supports the superiority of distributed practice compared to
massed practice on motor learning outcomes. However, inconsistencies in some findings claim for more efforts
on this topic.
AIM: We aimed to elucidate potential mechanisms that would support the distinct results between the different
distributed practice regimes (among days and within-one-day).
METHOD: Ninety participants, aged 18-25 years (45 men, 45 women) were randomly divided into 3 groups:
Massed practice group (MASSED) - 30 minutes of practice without rest/ one day; Within-one-day distributed
practice group (WITHIN) 6 blocks of 5 minutes of practice interspersed by 3 minutes of rest/ one day; and,
Among days distributed practice group (AMONG) 3 blocks of 10 minutes of practice divided into 3 consecutive
days. They practiced the three-ball cascade juggling task. The number of catches was the dependent variable.
There was a retention test (absolute retention and savings analyses) after 24 hours from the acquisition phase.
RESULTS: We identified that AMONG demonstrated higher absolute retention than WITHIN and MASSED.
Both distributed practice regimes demonstrated better savings than MASSED.
CONCLUSION: Our results showed that distributed practice regimes enhance consolidation processes and
information processing that benefit motor learning.
KEYWORDS: Practice distribution | Motor learning | Information processing | Consolidation | Savings
INTRODUCTION
Distributed practice refers to the spacing of practice over multiple sessions instead
of compacting practice time into a smaller number of sessions (massed practice)
1
. The
terms massed and distributed are defined in a continuum; typically, a massed practice
schedule involves longer active practice and shorter rest periods (or even without rest) in
comparison to a distributed schedule
2
.
Several studies have demonstrated that when practice amount is paired,
distributed practice tends to induce better motor performance and learning than massed
practice condition
3
. Interestingly, the superiority of the distributed practice compared to
massed practice has been reported in schedules in which practice is distributed within a
practice session (i.e., 30 minutes of massed practice versus 3 blocks of 10 minutes of
practice interspersed by 5 minutes of rest)
46
; and in distributed practice among days (i.e.,
30 minutes of massed practice versus 3 blocks of 10 minutes of practice interspersed into
3 consecutive days)
79
. In contrast, some authors attribute the inferiority of massed
practice in comparison to distributed practice only for motor performance instead of motor
learning outcomes (i.e., retention and transfer tests)
1012
. Yet, Lee and Genovese
13,14
identified that the distribution of practice enhances the learning of continuous motor skills,
BJMB
Brazilian Journal of Motor Behavior
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
154 of 161
Special issue:
The role of practice in motor learning
while massed practice schedules benefit the learning of discrete motor skills.
Despite these controversies, scientific literature generally attributes better motor
learning for distributed practice schedules. From a mechanistic perspective, it has been
suggested that distributed practice benefits motor learning by decreasing the deleterious
effect from fatigue, by increasing cognitive effort avoiding monotony and excessive
repetition, and by improving memory consolidation stimulating this mechanism across
several days (for the among days distributed practice)
2
.
The plausibility of all these mechanisms in explaining the effects of distributed
practice can be reexamined given that: 1 - even with fatigue the motor learning occurs
15
, 2
distributed practice does not necessarily engage a higher cognitive effort without some
attention direction, as visual instructions
1
, 3 motor memory consolidation is not
necessarily sleep-dependent (which would explain effects from distributed practice among
days). It has been suggested that motor memory consolidation also occurs during
wakefulness periods that are temporally close to practice
16
.
From these inconsistencies regarding the mechanisms involved in the superiority
of distributed practice compared to massed practice during motor learning, we seek to
explore whether the different regimes of distribution of practice (within-one-day and among
days) have different effects on motor learning. Given that possibly among days distributed
practice would influence motor learning through different mechanisms (motor memory
consolidation) if compared to within-one-day distributed practice. Some motor tasks
demonstrate to be more sensitive for sleep-dependent consolidation, while others seem to
have the consolidation process independent from sleep
17
. In our study, we used a three-
ball cascade juggling task that appears to have consolidation based on a sleep-dependent
mechanism
18,19
.
Based on three different practice regimes (massed practice, within-one-day
distributed practice, and among days distributed practice), we hypothesized the following
scenarios: In case of fatigue or cognitive effort is the main mechanism involved in
distributed practice phenomenon, both distributed practice regimes (within-one-day and
among-days) would demonstrate better retention than massed practice, without
differences between them; but, if consolidation is the main mechanism that benefits
distributed practice, among-days distributed practice would induce better retention than
within-one-day distributed practice and massed practice; however, if sleep, fatigue, and
cognitive effort have complementary effects on motor learning, both distributed practices
would demonstrate better retention than massed practice, but among-days distributed
practice would demonstrate superior retention than within-one-day distributed practice. To
the best of our knowledge, this is the first study investigating the effect of practice
distribution comparing the three distribution regimes (massed, within-one-day, and among
days).
METHODS
The ethics board from the State University of Piaui approved this study (protocol
number. 30456820.0.0000.5209). All participants signed the consent term before
participation. There was no compensation to participate in this study. All experiment was
conducted following Helsinki Declaration.
Participants
BJMB
Brazilian Journal of Motor Behavior
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
155 of 161
Special issue:
The role of practice in motor learning
We used the power analysis software G*Power 3.1
20
to determine the sample
size. Estimating an effect size of 0.66 (based on the performance difference across the
practice in Morita et al.
18
), α = 0.05, and a power of 0.8 indicated a sample size of 26 per
group.
We recruited 90 naïve participants from the local university community, aged 18-
25 years (21.62 ± 2.09); 45 men, 45 women. The inclusion criteria were: 1 To be a
university student. We adopted the following exclusion criteria: 1 To have cardiovascular
or osteoarticular diseases or disfunction which unviable the performance of the proposed
activities; 2 Do not have visual, neuromotor and cognitive conditions for understanding
and executing the proposed tasks; 3 - Do not use a corrective lens in case the participant
has unsatisfactory visual acuity; 4 To have previous experience in juggling tasks.
Apparatus and Task
We used three plastic balls with 76mm of diameter and mass of 160g for the
juggling task. As Morita et al.
18
performed in their study, participants threw the first ball
from the right to the left hand (the first catch). The second ball was thrown from the left to
the right hand (the second catch), then, if participants were able, the third ball was thrown
from the right to the left hand (the third catch). In this way, the task goal was to perform as
many catches as possible per trial without letting balls fall to the ground. When participants
failed to catch the balls, the experimenter delivered extra balls immediately to avoid delay
in restarting the subsequent trial. The dependent variable was the number of catches per
trial from the second catch. Thus, whether the participant performed five catches, we
computed three catches.
Procedure
After participants signed the consent form, they were divided into three groups
through a stratified random process considering sex: Massed Group (MASSED) (n = 30),
which completed the acquisition phase without rest (30 minutes of practice without rest/
one day); Within-one-day Distributed Practice (WITHIN) (n = 30), which had rest cycles
during the practice (6 x 5 minutes of practice interspersed by 3 minutes of rest/ one day);
and Among Days Distributed Practice (AMONG) (n = 30), (3 x 10 minutes of practice
divided into 3 consecutive days).
At the beginning of the experiment, each participant received an instructional
video about the motor pattern to perform the task. From the front view, the instructional
video demonstrated an experienced model performing the three-ball cascade juggling task
for 1 minute. The experient model performed 162 uninterrupted catches during the video.
After, participants were verbally instructed to initiate holding 2 balls in the right
hand, given that the first thrown should be performed from the right to the left hand. All
participants watched the same video and received the same verbal instruction. Then,
participants performed three trials as familiarization. They completed a pre-test consisting
of five trials (the duration of each trial depended on the number of catches that each
participant was able to achieve). After the pre-test, participants practiced the task for 30
minutes in the specific condition of their respective group (MASSED, WITHIN, or AMONG).
After the last trial of the acquisition phase, participants rested for 2 minutes, and then they
performed a post-test with the same pre-test characteristics.
After 24 hours from the post-test, participants completed a retention test with the
BJMB
Brazilian Journal of Motor Behavior
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
156 of 161
Special issue:
The role of practice in motor learning
same pre-test and post-test characteristics. Following, participants of all groups completed
30 minutes of massed practice to verify whether they demonstrated savings.
Statistical Analyses
We used STATISTICA 11.0 (StatSoft Inc., Tulsa, OK, USA) and Microsoft Excel
365 softwares for statistical analyses adopting a 5% significance level. We evaluated the
normality and homogeneity of the data with the Shapiro Wilks and Levene tests,
respectively.
We assessed the motor performance changes using the mean of catches in each
block of trial (pre-test, post-test, and retention test). As inferential analyses, we performed
an ANOVA two-way - 3 groups (MASSED, WITHIN, AMONG) x 3 times (pre-test, post-test,
retention test) with repeated measures in the second factor. Fisher test was used for post
hoc analyses.
As Schmidt et al.
21
stated, savings is a measure of retention which involves the
rate of relearning; “that is, after a retention interval, one measures the number of trials
required for the participants to reach the level of proficiency achieved in original practice”.
We evaluated savings by computing the number of trials (on the second day) required for
participants to reach the mean performance achieved in the post-test. Then, we verified
the effects of the practice distribution on savings, comparing the number of trials among
MASSED, WITHIN, and AMONG through an ANOVA one-way with a Fisher post hoc test.
We reported Cohen’s d as an estimate for effect sizes.
RESULTS
Considering the mean of catches during acquisition phase and retention test, we
identified an interaction effect in the ANOVA two-way (F
2,87
= 5.86, p < .0001, d = .66). The
Fisher post hoc test demonstrated that there was no difference among groups in pre-test
and post-test, and all of them improved their performance comparing pre-test and post-test
(MASSED, p < .0001; WITHIN, p < .0001; AMONG, p < .0001). There was no significant
difference between post-test and retention test for all groups (MASSED, p= .17; WITHIN,
p = .22; AMONG, p = .76), which indicates that all groups learned the juggling task.
However, AMONG demonstrated significant difference in retention test if compared to
MASSED (p = .01) and WITHIN (p = .01); there were no significant difference between
MASSED and WITHIN in retention test (p = .89). These results can be checked in Figure
1.
BJMB
Brazilian Journal of Motor Behavior
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
157 of 161
Special issue:
The role of practice in motor learning
Figure 1. Mean of catches for WITHIN, MASSED, and AMONG across acquisition phase and retention. Data
represent mean and confidence interval 95%.
In our savings analysis, the ANOVA one-way demonstrated a significance in group
factor (F
2,87
= 3.59, p < .05, d = .6). The Fisher post hoc test demonstrated a significant
difference between AMONG and MASSED (p = .01) and WITHIN and MASSED (p < .05).
Interestingly, there was no significant difference between AMONG and WITHIN (p = .89).
In this way, it is possible to verify that AMONG and WITHIN needed fewer trials in savings
to achieve the mean performance of the post-test than MASSED. These findings can be
verified in Figure 2.
Figure 2. Blocks of trials on day 2 to achieve the post-test mean performance. Data show the mean of blocks
in each group (solid line) and individual data (dots)
BJMB
Brazilian Journal of Motor Behavior
Luz, Santos,
Bonuzzi
2022
VOL.16
N.2
158 of 161
Special issue:
The role of practice in motor learning
DISCUSSION
We aimed to investigate the effects of two different distributions of practice (with-
in-one-day and among days) in motor learning. Based on possible mechanisms that these
different practice distributions could engage, we hypothesized that they could affect motor
learning differently. The literature has attributed for the distributed practice the higher
engagement for cognitive processing and lower monotony, as well it has been postulated
that among-days distributed practice engages more consolidation mechanisms which
could induce better retention
2
. Our findings support the previously mentioned
mechanisms. We found that among-days distributed practice influenced consolidation
mechanisms that allow better absolute retention (first trials of a retention test) than within-
one-day distributed practice and massed practice. However, both distributed regimes
demonstrated faster savings than massed practice, without differences between them.
Practice distribution can be considered as one of the classic themes in the motor
learning field. Previous studies have already demonstrated the superiority of distributed
practice compared to massed practice
49
. However, part of the literature advocates that
distributed practice has little or no effect on learning and concludes that massing of
practice impairs just motor performance instead of any permanent decremental states
(motor learning)
1012
. Based on our experimental design, we can suppose the
mechanisms involved in the superiority of distributed practice regimes compared to
massed practice: motor memory consolidation and higher cognitive engagement.
Consolidation is the post-practice phase when the memory becomes more robust
and stable with less susceptibility to interference
22,23
. Our results corroborate the
hypothesis that among days distributed practice can impact motor learning enhancing
consolidation processes. Given that, the maintenance of improved performance in
absolute retention (first trials of a retention test) reveals the effectiveness of a
consolidation process of a particular motor memory
24
. So, we attributed the fact that
AMONG had superior absolute retention than WITHIN and MASSED by the prolonged
influence of this practice distribution on consolidation processes. Due to the AMONG
regime, the participants had a multi-day consolidation process (3 consecutive days of
sleep-dependent consolidation) that did not occur in WITHIN or MASSED participants, with
only 1 sleep-dependent consolidation event.
Several findings suggest that motor memory consolidation occurs temporally close
to practice (in wakefulness) and during sleep
16
. Additionally, the participation of sleep-
dependent or wakefulness-dependent consolidation processes seems to differ according
to the task nature or demand
23
. For example, it has been suggested that movement
components have consolidation processes during wakefulness, while the goal component
of the motor skill is consolidated in a sleep-dependent condition
23
. Thus, the influence of
the within-one-day distributed practice on consolidation may depend on whether the task
to be learned had the consolidation process engaged in a wakefulness-dependent
condition. In our case, the juggling task seems to have the consolidation process in a
sleep-dependent condition
18
, which explains the superiority of the AMONG in the absolute
retention in our study.
In fact, the interaction between distribution practice regime, type of motor task and
consolidation mechanisms (sleep-dependent or wakefulness-dependent) may explain that
in some cases, the distributed practice did not demonstrate superiority to massed practice