Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Sep 11;64(5):e70678. doi: 10.1111/ejn.70678

The Effects of Anxiety and Cognitive Load on Freezing of Gait in Parkinson's Disease

Demi Zoetewei 1, Moran Gilat 1, Nicholas D'Cruz 1, Jan Cornelis 2, Jordan Bedime Songue Njiki 3, Rick C Helmich 4, Pieter Ginis 1,5, Alice Nieuwboer 1,✉
PMCID: PMC13564402  PMID: 42725995

ABSTRACT

It is currently unclear whether anxiety, cognitive load, or their combination modulates freezing of gait differently and whether people with Parkinson's disease and freezing of gait (freezers) suffer from stronger threat responsiveness than nonfreezing groups. Therefore, we investigated 28 freezers, 19 nonfreezers, and 19 healthy age‐matched controls, who completed a freezing‐provoking protocol under different conditions. State anxiety was generated via a threat of shock or a threat of freezing paradigm. Cognitive load was imposed via a dual task. The primary outcome was the freezing ratio, derived from wearable sensors. Stress was monitored by self‐reported and physiological markers. Freezers had worse cognition and more severe motor symptoms than nonfreezers yet a similar trait anxiety and threat response compared with both nonfreezing groups. Anxiety and cognitive load tended to increase the freezing ratio across groups. When combined, anxiety and cognitive load showed no surplus effects. Compared with the threat of shock, the threat of freezing had less impact on the freezing ratio, and although it induced less perceived stress, it generated higher physiological stress responses. In conclusion, freezers showed overall comparable levels of trait anxiety and a similar threat response as nonfreezers and healthy controls. Anxiety and cognitive load each tended to increase the freezing ratio, but their combined effect seemed to plateau. The threat of freezing had a smaller effect than the threat of shock as the former may not have diverted the attention away from walking to the same degree as the latter, although this explanation remains speculative.

Keywords: anxiety, cognitive load, freezing of gait, Parkinson's disease, threat of shock

During a FOG‐provoking protocol, a threat of receiving an annoying shock and a cognitive load both worsened the FOGratio but combining them did not have a surplus effect. This suggests a complex modulation between anxiety and cognitive load. Interestingly, a threat of an upcoming freezing episode did not worsen FOG, possibly because a threat of shock may distract from gait whereas a threat of freezing may emphasize attention on gait. Effects of anxiety were similar between freezers and control groups.

graphic file with name EJN-64-0-g002.webp

Abbreviations

%TF

percentage time frozen

Baseline

condition in single task without threat

BDI

Beck Depression Inventory

Controls

people without Parkinson's disease

DT + TOS

dual task with threat of shock

DT

dual task

ECG

electrocardiogram

FOG

freezing of gait

FOGratio

freezing ratio

Freezers

patients with Parkinson's disease and freezing of gait

GSR

galvanic skin response

H&Y

Hoehn and Yahr

HR

heart rate

LEDD

Levodopa Equivalent Daily Dose

MDS‐UPDRS

Movement Disorders Society—Unified Parkinson's Disease Rating Scale

MoCA

Montreal Cognitive Assessment

NFOG‐Q

New Freezing of Gait Questionnaire

Nonfreezer

patients with Parkinson's disease without freezing of gait

ON medication

with the regular intake of antiparkinsonian medication

PAS

Parkinson Anxiety Scale

PD

Parkinson's disease

SCOPA‐AUT

Scales for Outcomes in Parkinson's Disease—Autonomic Dysfunction

ST

single task

TOF

threat of freezing

TOS

threat of shock

VAS

Visual Analog Scale

1. Introduction

Freezing of gait (FOG) episodes, defined as “paroxysmal episodes wherein there is an inability to step effectively despite attempting to do so” (Gilat et al. 2026) are very disabling for people with Parkinson's disease (PD). The fact that FOG is unpredictable and strongly associated with falls (Bloem et al. 2004) gives rise to frustration, anxiety, and fear of falling (Ehgoetz Martens et al. 2020). Furthermore, trait anxiety (the tendency to experience anxiety across situations) was found to be a predictor for the emergence of FOG (Ehgoetz Martens et al. 2018), and state anxiety (the transient limbic response to a perceived threat) was found to exacerbate the severity of FOG, particularly under complex conditions (Economou et al. 2021; Ehgoetz Martens et al. 2014, 2015). The latter has led to the hypothesis that cross‐talk between limbic, motor, and cognitive circuits may overload impaired striatal capacity and induce FOG (Shine et al. 2011), further supported by associations between FOG severity and dysfunctional limbic circuitry (Gilat et al. 2018) and noradrenergic activity (Cockx et al. 2023; Economou et al. 2021; Maidan et al. 2010; Mazilu et al. 2015; Taylor et al. 2022). Even in individuals without freezing, adding a dual task (DT) to performing ankle movements in the scanner was associated with a larger area of striatal brain activity than in healthy controls, suggesting reduced striatal focusing in PD (Nieuwhof et al. 2017). It is also known that people with PD are overly reliant on attentional control to regulate their walking due to motor automaticity failure (Vandenbossche et al. 2012) and that anxiety may direct attention towards threat‐related stimuli (attentional control theory) (Eysenck et al. 2007). In addition, freezers have reported various FOG‐related threats in daily life, such as fear of another episode and fear of falls (Duysens and Smits‐Engelsman 2023) and may thus be oversensitive to state anxiety (Duysens and Smits‐Engelsman 2023; Gilat et al. 2018). However, how state anxiety and cognitive load interact to modulate FOG remains poorly understood. It is also unknown whether FOG is affected differently by FOG‐specific versus nonspecific threats.

Here, we manipulated anxiety and cognitive load, both alone and together, to investigate their impact on FOG severity in freezers, nonfreezers, and healthy controls during a FOG‐provoking protocol. The primary outcome was the freezing ratio (FOGratio), derived from inertial measurement units and reflecting FOG‐related gait breakdown, which allowed for comparison across PD nonfreezers and healthy controls (Mancini et al. 2012). To evaluate cognitive load, we used single task (ST) and DT conditions. To provoke state anxiety, we applied a classic “threat of shock” (TOS) paradigm (Schmitz and Grillon 2012). We hypothesized that increasing limbic (TOS) and cognitive (DT) load would worsen the FOGratio, particularly when combined (DT + TOS), and that this effect would be stronger in freezers than in the nonfreezing groups (Shine et al. 2011). We also explored the effects of a threat of an upcoming FOG episode (threat of freezing [TOF]) using a novel sham paradigm.

2. Materials and Methods

2.1. Study Design

This cross‐sectional study, with a within‐ and between‐factor design, was preregistered at Open Science Framework Registries (osf.io/kh5ny). The full protocol was approved by the ethical committee of UZ Leuven (S62453). Participants signed an informed consent prior to the study‐related procedures.

2.2. Participants

A total of 28 PD freezers, 19 PD nonfreezers, and 20 age‐matched healthy controls were recruited through the contact database of the Parkinson Rehabilitation Research Lab (pro‐labo.be) at KU Leuven (for details sample size, see osf.io/kh5ny). Inclusion criteria were age > 40 years, the ability to walk for 5 min, a score of ≥ 24 on the Mini‐Mental State Examination (Folstein et al. 1975), and a modified Hoehn and Yahr (H&Y) stage of I–IV in the ON state (Hoehn and Yahr 1967; Jankovic et al. 1990). Participants were excluded if they had severe orthostatic hypotension (≥ 3 on Item 1.12 of the Movement Disorders Society—Unified Parkinson's Disease Rating Scale [MDS‐UPDRS]) (Goetz et al. 2008), a severe skin allergy, a pacemaker, a deep brain stimulator, and comorbidity impacting mobility and/or mental health. Clinical diagnosis of idiopathic PD was based on the UK PD Society Brain Bank criteria (Hughes et al. 1992). PD patients were classified as freezers if they had a score > 0 on the New Freezing of Gait Questionnaire (NFOG‐Q) (Nieuwboer et al. 2009) and/or if FOG was clinically observed.

2.3. Study Procedures

The experiment took place in the practical ON state, 1–2 h after intake of the regular dose of antiparkinsonian medication. A short eligibility and cognitive battery was performed first. The remaining clinical data were collected after the main experiments, including the Levodopa Equivalent Daily Dose (LEDD) (Jost et al. 2023), the NFOG‐Q, the Parkinson Anxiety Scale (PAS) (Leentjens et al. 2014), the Beck Depression Inventory (BDI) (Beck et al. 1961), the MDS‐UPDRS, and the Scales for Outcomes in Parkinson's Disease—Autonomic Dysfunction (SCOPA‐AUT) (Visser et al. 2004).

Figure 1 displays the shortened version of a previously validated FOG‐provoking gait protocol (Ziegler et al. 2010), including standing for 10 s, walking 2 m, performing four alternating 360° turns within a square (40 × 40 cm), walking another 2 m diagonally and opening a door, performing a 180° turn within the doorway, and returning to the starting position. Participants were instructed to perform the task as fast and as safely as possible. The protocol was performed in five conditions: (1) baseline (ST without threat), (2) DT, (3) TOS, (4) DT + TOS, and (5) TOF. Each condition consisted of three trials, except for the TOF, which only included two trials. The order of conditions was counterbalanced within‐group except for the baseline (first) and the TOF (last). A video camera and Opal inertial measurement units, attached to the feet, shins, and lower back (APDM Inc., Portland, OR, USA), were used to record walking. The wearable Chill Band at the wrist and the Health Patch at the sternum (Imec, Leuven, Belgium) recorded galvanic skin responses (GSRs) and electrocardiography (ECG), respectively (Smets et al. 2019). A harness fitted on the back (1.72 kg) included custom trigger boxes for synchronization. After each condition, participants indicated their perceived stress levels on a Visual Analog Scale (VAS).

FIGURE 1.

FIGURE 1

Open in a new tab

FOG‐provoking protocol. The task consisted of standing for 10 s, walking 2 m, performing 4 x 360° turns in place with alternating directions, walking towards and opening a door, turning 180° in the doorway, and returning to the start position. The task was performed 3× in all conditions, except for the ST TOF condition which consisted of two trials. Baseline: without cognitive load and threat manipulation; DT: dual task (auditory Stroop task); TOF: threat of FOG; TOS: threat of shock.

The DT consisted of the auditory Stroop task (Kestens et al. 2021), whereby participants heard the words “high” and “low” spoken congruently or incongruently in either a high or a low pitch through a headset (Figure 2A). Subjects were instructed to verbally report the pitch of each word and were familiarized with this task prior to walking.

FIGURE 2.

FIGURE 2

Open in a new tab

(A) Setup during the FOG‐provoking protocol, (B) electrodes for shock delivery during TOS, and (C) custom‐made braces for TOF. The electrodes and braces were worn during the full protocol but only used during the TOS or TOF condition, respectively. During TOF, participants were told that the braces could be turned on at any time and that this would induce or worsen FOG. To convince participants, they were shown an edited video of a patient experiencing severe FOG when the braces were activated.

The TOS delivered to the index finger during walking was used to induce state anxiety (Figure 2A,B) (McDonald et al. 2015; Robinson et al. 2013; Schmitz and Grillon 2012; van Langen 2019). First, shock intensity was individually calibrated as annoying but not painful, indicated by a score of 4 out of 5 (for details, see Data S1) (Klumpers et al. 2010). If a score of 4 was not achieved, the maximum intensity was used. Participants were then informed that shocks could occur at any moment during the walking trials. Based on pilot testing of the current protocol, shocks were administered at a higher frequency than in previous studies to maintain anticipation (zero, one, or two shocks per trial) (McDonald et al. 2015; Schmitz and Grillon 2012; van Langen 2019). Shock timing was predefined and differed between the TOS and DT + TOS conditions. Shocks were delivered using rubber electrodes attached with conductive adhesive gel (Innostim TENS 2000, MediMaxTech, New Malden, UK). The stimulus consisted of 25 pulses with a pulse width of 250 μs, a frequency of 100 Hz, and a personalized intensity between 16 and 80 mA.

To induce fear of an impending FOG episode (TOF), participants wore custom‐built sham FOG‐provoking leg braces (Figure 2A,C) which had no genuine FOG‐inducing mechanism and were used purely to create a credible threat. Before the trial, they were informed that the braces could be activated to induce FOG, even in nonfreezers and controls, illustrated by a video showing severe FOG during brace activation accompanied by vibration and flashing lights (Video S1). Only the first two trials, in which the braces were not activated but FOG was anticipated, were analyzed. In two subsequent trials, the braces were activated before and after participants were informed that they did not actually provoke FOG (not reported).

2.4. Outcome Measures

The primary outcome, the FOGratio, was extensively validated in a previous study (Mancini et al. 2012). Higher values indicate worse FOG‐related stepping breakdown (D'Cruz et al. 2022; Mancini et al. 2017, 2023) (for details, see Data S2). Secondary outcomes were the FOGratio during the 360° turns only (the most sensitive trigger of FOG; D'Cruz et al. 2022; Zoetewei et al. 2024), the duration of the FOG‐provoking protocol, video‐based FOG severity, the perceived stress, and the physiological stress measures. Start and end time of the 360° turns were determined automatically from the APDM system (El‐Gohary et al. 2014) and checked visually. The percentage time frozen (%TF) and the number of FOG episodes were determined from video annotations by a trained tester using Elan software (ELAN [Version 5.5] 2019; Gilat 2019) (for definitions, see Data S3). Physiological stress measures included the mean heart rate (HR) and mean GSR. All outcomes were calculated for each trial and averaged per condition, except perceived stress which already consisted of one value per condition. Data processing was performed using MATLAB, Version R2020b (MathWorks, Natick, MA, USA).

2.5. Statistical Analysis

Statistical analyses were performed using the Statistical Package for the Social Sciences, Version 28 (IBM SPSS Statistics, Chicago, IL, USA) and the SAS software, Version 9.4 for Windows (SAS Institute Inc., Cary, NC, USA). Alpha was set at 0.05. Deviations from the OSF registration are described in Data S4.

Depending on data normality (Shapiro–Wilk test) and the type of outcome (continuous/categorical), descriptives were reported as mean (SD), median (IQR) or frequencies, and tested for group differences parametrically using a one‐way ANOVA with Tukey HSD post hoc tests or independent t‐test or nonparametrically using a Kruskal–Wallis H or Mann–Whitney U test.

In Model 1, we evaluated the effects of the following independent variables and their interactions: anxiety (no threat/TOS), cognitive load (ST/DT), and group (freezers/nonfreezers/controls). In Model 2, we explored differences between TOS and TOF exposures, including the variables anxiety (no threat/TOS/TOF), group, and anxiety * group. This separation into two analyses was motivated by statistical parsimony and by the fact that TOF was exploratory and not administered in combination with the DT. We used the same models, excluding group and interactions, to analyze the video‐rated FOG outcomes, only available in freezers. For the sensitivity analysis, we excluded people who did not perceive the shock as sufficiently annoying, based on a score of < 4 out of 5.

Linear mixed models (LMMs) were evaluated using the MIXED procedure in SAS with the restricted maximum likelihood fitting criterion. Continuous variables were transformed with an inverse hyperbolic sine, which improved the distribution of the residuals. They were analyzed with LMMs with an unstructured covariance structure. We evaluated the number of FOG episodes, a count variable, using a generalized LMM with a log link function, a Poisson distribution, and a compound symmetry covariance structure. Cohen's d absolute effect sizes were calculated using the Satterthwaite degrees of freedom and t‐statistics (Lakens 2013). Post hoc results were adjusted for multiple comparisons using Tukey–Kramer corrections.

3. Results

3.1. Participants

One healthy participant dropped out due to an emotional response to the TOS condition, resulting in a sample size of 28 freezers, 19 nonfreezers, and 19 controls (Table 1). During the protocol, 24 freezers showed FOG. However, only nine freezers had an average %TF > 5%. In seven severe freezers (NFOG‐Q > 16), the protocol was shortened for reasons of feasibility. For the sensitivity analysis, seven freezers, four nonfreezers, and two controls were excluded because they did not perceive the shock from the TOS protocol as sufficiently annoying. Due to data loss, HR and GSR were available for 20 and 28 freezers, 18 and 19 nonfreezers, and 17 and 16 controls, respectively.

TABLE 1.

Descriptive data per group.

Freezers (n = 28) Nonfreezers (n = 19) Controls (n = 19) p
Age a 66.0 (9.00) 71.0 (7.00) 67.0 (12.0) p = 0.520
MoCA a 25.0 (4.00) 27.0 (3.00) 27.0 (3.00) p = 0.004*, b , c
NFOG‐Q 15.3 (7.64) 0.00 (0.00) 0.00 (0.00) p < 0.001*, b , c
PAS 9.96 (5.56) 10.6 (7.33) 6.68 (4.92) p = 0.096
BDI a 11.0 (7.00) 7.00 (7.00) 4.00 (3.00) p < 0.001*, ,c
SCOPA‐AUT 21.7 (10.4) 17.5 (9.72) 8.78 (4.37) p < 0.001*, c , d
Disease duration 11.3 (6.29) 6.79 (3.87) n/a p = 0.004*
LEDD a 924 (580) 680 (560) n/a p = 0.004*
MDS‐UPDRS I a 10.5 (6.00) 8.00 (7.00) n/a p = 0.128
MDS‐UPDRS II a 17.0 (7.00) 9.00 (5.00) n/a p < 0.001*
MDS‐UPDRS III 41.3 (11.3) 36.0 (10.7) n/a p = 0.109
MDS‐UPDRS IV a 3.50 (8.00) 1.00 (8.00) n/a p = 0.308
MDS‐UPDRS total a 73.5 (29.0) 56.0 (23.0) n/a p = 0.009*
H&Y, 2/2.5/3/4 a 7/15/5/1 11/5/3/0 n/a p = 0.057
Open in a new tab

Note: Values are reported as mean (SD) and were tested for group differences using a one‐way ANOVA with Tukey HSD post hoc tests or independent t‐test.

Abbreviations: BDI: Beck Depression Inventory; H&Y: Hoehn and Yahr (modified); LEDD: Levodopa Equivalent Daily Dose; MDS‐UPDRS: Movement Disorders Society—Unified Parkinson's Disease Rating Scale; MoCA: Montreal Cognitive Assessment; NFOG‐Q: New Freezing of Gait Questionnaire; PAS: Parkinson Anxiety Scale; SCOPA‐AUT: Scales for Outcomes in Parkinson's Disease—Autonomic Dysfunction.

a

Values are reported as median (IQR) or frequencies and were tested for group differences using a Kruskal–Wallis H or Mann–Whitney U test.

b

Significant difference between freezers and nonfreezers.

c

Significant difference between freezers and controls.

d

Significant difference between nonfreezers and controls.

*

Significantly different (p < 0.05).

Table 1 shows that on average, freezers had worse Montreal Cognitive Assessment (MoCA) scores than nonfreezers (p = 0.038) and controls (p = 0.007) and higher depression than controls (p < 0.001). Both freezers (p < 0.001) and nonfreezers (p = 0.012) had more impaired autonomic functioning compared with controls. Additionally, freezers had a higher LEDD (p = 0.004), a tendency to higher H&Y stages, worse MDS‐UPDRSII (p < 0.001), and MDS‐UPDRS total scores (p = 0.009) than nonfreezers. Of note, trait anxiety (PAS scores) only showed a trend towards a group effect (p = 0.096), mostly driven by differences between PD and controls (p ≥ 0.117).

3.2. Group Effects and Interactions

Group effects for Model 1 (in relation to TOS and DT) are displayed in Table 2 and Figure 3 and show that freezers exhibited higher FOGratios during the FOG‐provoking protocol compared with both nonfreezers (d = 0.68, p = 0.021) and controls (d = 0.96, p < 0.001). Values did not differ significantly between nonfreezers and controls (p = 0.307) (for more information on data variability, see Data S5). These findings were similar for 360° turns, for task duration and in the sensitivity analysis. However, when evaluating the type of threat (Model 2: TOS vs. TOF) as shown in Figure 4 and Table S6.1, a significantly higher FOGratio and longer task duration during the full protocol were found for freezers compared with controls only. There were no significant group differences for any of the physiological stress measures, unlike perceived stress (d = 0.68, p = 0.001). Post hoc, freezers reported higher stress levels than controls (d = 0.97, p < 0.001) but not compared with nonfreezers (p = 0.190).

TABLE 2.

The effects of anxiety and cognitive load on FOG‐related outcomes between the groups.

Data means (SD) Results F‐tests LMMs
Baseline DT TOS DT + TOS Group Anxiety Cognitive load Anxiety * cognitive load
Freezers, nonfreezers, and controls
Freezing ratio Freezers 1.49 (1.09) 1.90 (1.83) 1.59 (1.04) 1.51 (0.97) Cohen's d d = 0.71 d = 0.07 d = 0.23 d = 0.26
Nonfreezers 0.87 (0.52) 0.95 (0.51) 0.99 (0.55) 0.96 (0.50) p p < 0.001 p = 0.565 p = 0.068 p = 0.045
Controls 0.71 (0.34) 0.76 (0.45) 0.76 (0.50) 0.76 (0.40) Significance 1 a , b , 2 a , b
Freezing ratio 360° turns Freezers 3.20 (2.79) 3.81 (4.57) 3.69 (3.21) 3.11 (3.00) Cohen's d d = 0.91 d = 0.28 d = 0.10 d = 0.24
Nonfreezers 1.30 (1.04) 1.74 (1.76) 1.72 (1.48) 1.77 (1.64) p p < 0.001 p = 0.029 p = 0.430 p = 0.059
Controls 0.77 (0.57) 0.85 (0.79) 0.89 (0.69) 0.91 (0.65) Significance 1 a , b , 2 a 1, 2 2 c
Task duration (s) Freezers 62.6 (49.5) 71.6 (77.0) 55.6 (47.1) 66.5 (69.5) Cohen's d d = 0.95 d = 0.40 d = 0.16 d = 0.22
Nonfreezers 35.0 (6.66) 33.4 (7.43) 32.2 (7.12) 33.6 (7.09) p p < 0.001 p = 0.002 p = 0.214 p = 0.081
Controls 28.2 (5.29) 26.9 (4.70) 25.6 (2.99) 26.5 (4.24) Significance 1 a , b , 2 a , b 1, 2
Freezers only
%TF (%) Freezers 13.5 (25.3) 16.3 (27.8) 15.0 (26.2) 14.8 (26.1) Cohen's d n/a d = 0.02 d = 0.10 d = 0.08
p n/a p = 0.916 p = 0.608 p = 0.674
N FOG (#) Freezers 2.20 (4.05) 2.59 (4.40) 1.56 (1.97) 2.32 (4.21) Cohen's d n/a d = 0.13 d = 0.16 d = 0.07
p n/a p = 0.229 p = 0.146 p = 0.534
Open in a new tab

Note: Interactions with group were not significant and are not shown here. Significance (alpha = 0.05) is indicated for the main and sensitivity analysis and for post hoc results: 1: significant effect in main analysis; 2: significant effect in sensitivity analysis.

Abbreviations: %TF: percentage time frozen; Baseline: without cognitive load and threat; DT: dual task; LMMs: linear mixed models; N FOG: the number of FOG episodes; SD: standard deviation; TOS: threat of shock.

a

Significant difference between freezers and controls.

b

Significant difference between freezers and nonfreezers.

c

Significant difference between baseline and TOS.

FIGURE 3.

FIGURE 3

Open in a new tab

The FOGratio during cognitive load (DT) and anxiety (TOS) in freezers, nonfreezers, and controls. Baseline: without cognitive load and threat; DT: dual task; TOS: threat of shock. Bars and error bars represent means and standard deviations, respectively. Indicated by *, the FOGratio was significantly higher in freezers than in nonfreezers (p = 0.021) and controls (p < 0.001). Across groups, the FOGratio was higher during DT, TOS, and DT + TOS compared with baseline, albeit not significantly after multiple comparison correction.

FIGURE 4.

FIGURE 4

Open in a new tab

The FOGratio during baseline TOS and TOF in freezers, nonfreezers, and controls. Baseline: without threat; TOF: threat of FOG; TOS: threat of shock. Bars and error bars represent means and standard deviations, respectively. As indicated with *, the FOGratio was significantly higher in freezers than controls (p = 0.003) and higher during TOS compared with baseline across groups.

No significant interactions for group with anxiety or cognitive load were found (in both models). Condition effects are therefore described across groups.

3.3. Model 1: Effects of Anxiety (TOS) and Cognitive Load

For the primary outcome, we found a significant anxiety * cognitive load effect (d = 0.26, p = 0.045), requiring post hoc testing. Compared with baseline, the FOGratio was increased to a similar degree by adding a TOS (d = 0.32, p = 0.065), DT (d = 0.27, p = 0.149), or DT + TOS (d = 0.28, p = 0.121), albeit not significantly after multiple comparison correction. No significant differences were found between DT, TOS, and DT + TOS. Mean differences were largely similar for the FOGratio during 360° turns (Table 2). However, during 360° turns, possibly because of high standard deviations, significant effects were found for anxiety (d = 0.28, p = 0.029) but not for cognitive load (d = 0.10, p = 0.430). In the sensitivity analysis, anxiety * cognitive load was significant (d = 0.29, p = 0.044), with post hoc results revealing higher values during TOS compared with baseline (d = 0.45, p = 0.012). Video‐rated FOG severity (in freezers only) was not affected by anxiety, cognitive load, or their interaction. Interestingly, while FOG‐related outcomes tended to be increased in response to the various load conditions, the duration to complete the FOG‐provoking protocol was significantly lowered by anxiety (d = 0.40, p = 0.002) (Figure 5).

FIGURE 5.

FIGURE 5

Open in a new tab

Heart rate in relation to different types of threats (baseline, TOS, and TOF) in freezers, nonfreezers, and controls. Baseline: without threat; TOF: threat of FOG; TOS: threat of shock. Bars and error bars represent means and standard deviations, respectively. Across groups, HR was higher during TOF compared with baseline and TOS, indicated with *.

3.4. Model 2: Effects of Type of Threat (TOS vs. TOF)

Significant effects of the type of threat were found for the FOGratio (d = 0.24, p = 0.032) and the FOGratio during 360° turns (d = 0.29, p = 0.008) (see Table S6.1). Remarkably, post hoc comparisons showed that only the TOS, and not the TOF, increased the FOGratio compared with baseline. Both types of threat reduced the task duration (TOS: d = 0.70, p < 0.001; TOF: d = 0.39, p = 0.007). The sensitivity analysis showed a significant effect of type of threat on %TF (d = 0.46, p = 0.025), revealing a reduction of %TF in the TOF compared with the TOS condition (d = 0.59, p = 0.034), but this was not observed for the number of FOG episodes (d = 0.17, p = 0.182).

3.5. Effects on Physiological and Perceived Stress

HR was not influenced by either anxiety or cognitive load (Table 3). In contrast, the GSR showed a significant interaction between these factors (d = 0.59, p < 0.0001). Post hoc analysis revealed that GSR levels were significantly elevated relative to baseline during the TOS (d = 0.71, p < 0.001), DT (d = 0.64, p < 0.001), and DT + TOS (d = 0.68, p < 0.001). Perceived stress showed significant main effects of both anxiety and cognitive load, indicating that combining the two threats produced a surplus effect, rather than an interaction‐driven increase (anxiety * cognitive load p = 0.332).

TABLE 3.

The effects of anxiety and cognitive load on stress‐related outcome measures between the groups.

Data means (SD) Results F‐tests LMMs
Baseline DT TOS DT + TOS Group Anxiety Cognitive load Anxiety * cognitive load
HR Freezers 80.4 (8.91) 79.9 (9.27) 81.1 (9.64) 81.3 (10.1) Cohen's d d = 0.17 d = 0.16 d = 0.07 d = 0.17
Nonfreezers 82.9 (13.7) 83.7 (14.1) 82.3 (13.1) 84.7 (15.4) p p = 0.694 p = 0.243 p = 0.624 p = 0.224
Controls 84.9 (13.9) 83.7 (12.0) 84.3 (12.1) 84.3 (12.4) Significance
GSR Freezers 2.58 (4.16) 3.50 (4.80) 3.08 (4.35) 3.30 (4.62) Cohen's d d = 0.12 d = 0.44 d = 0.50 d = 0.59
Nonfreezers 2.11 (3.99) 2.57 (4.23) 2.58 (4.22) 2.68 (4.36) p p = 0.810 p = 0.001 p < 0.001 p < 0.001
Controls 2.55 (4.19) 3.18 (4.96) 3.35 (5.05) 3.31 (5.20) Significance 1, 2 1, 2 1 c , d , e , 2 c , d , e
Perceived stress (VAS 0–10) Freezers 3.79 (2.11) 5.11 (2.67) 4.41 (2.42) 5.44 (2.51) Cohen's d d = 0.68 d = 0.38 d = 1.06 d = 0.12
Nonfreezers 2.30 (1.43) 4.43 (1.81) 3.19 (2.21) 4.56 (2.23) p p = 0.001 p = 0.004 p < 0.001 p = 0.332
Controls 1.79 (1.94) 2.49 (2.67) 2.04 (1.92) 3.03 (2.45) Significance 1 a , 2 a 1, 2 1, 2
Open in a new tab

Note: Interactions with group were not significant and are not shown here. Significance (alpha = 0.05) is indicated for the main and sensitivity analysis and for post hoc results: 1: significant effect in main analysis; 2: significant effect in sensitivity analysis.

Abbreviations: Baseline: without cognitive load and threat; DT: dual task; LMMs: linear mixed models; SD: standard deviation; TOS: threat of shock; VAS: Visual Analog Scale (a continuous scale marked on a horizontal line).

a

Significant difference between freezers and controls.

b

Significant difference between freezers and nonfreezers.

c

Significant difference between Baseline and TOS.

d

Significant difference between Baseline and DT.

e

Significant difference between Baseline and DT + TOS.

Regarding the different types of threat (Model 2), both TOS and TOF provoked significant increases of HR (d = 0.32, p = 0.006), GSR (d = 0.58, p < 0.001), and perceived stress (d = 0.28, p = 0.011; see Table S6.2). Perceived stress was higher in TOS compared with both baseline (d = 0.35, p = 0.019) and TOF (d = 0.33, p = 0.029), with no difference between baseline and TOF. Conversely, the physiological measures indicated greater stress responses in the TOF than in TOS, reflected by higher HR (d = 0.46, p = 0.005) and GSR (d = 0.50, p = 0.010). GSR increased significantly from baseline to both TOS and TOF, whereas HR was significantly higher from baseline to TOF (d = 0.34, p = 0.046) but did not differ between baseline and TOS.

4. Discussion

In this study, we directly compared the effects of two experimental manipulations of anxiety and cognitive load on gait breakdown during a FOG‐provoking protocol in freezers, nonfreezers, and healthy controls. Physiological and perceived stress markers verified that stress levels were effectively modulated. Imposing either anxiety or cognitive load, whether through a TOS or a DT, resulted in a tendency towards greater gait breakdown, with small to medium effect sizes that were largely comparable across groups. Contrary to our expectations, the TOS + DT condition did not further increase FOGratios nor affect other FOG‐related outcomes. Likewise, the TOF did not worsen these outcomes; rather, it led to an improvement in %TF compared with TOS. Notably, TOF elicited lower perceived stress, yet it was accompanied by higher physiological stress. Trait anxiety and physiological stress were comparable across groups, except for perceived stress immediately following the conditions, which was higher in freezers than in healthy controls.

FOG has been associated with increased cross‐talk between motor, cognitive, and limbic neural circuits, superimposed on reduced processing capacity, especially during periods of dopamine deficiency (Lewis and Shine 2016; Shine et al. 2011). Accordingly, we anticipated that combining anxiety and cognitive load would exacerbate FOG (Shine et al. 2011). However, this interaction did not produce an accumulative effect in our data collected during ON medication. Hence, future studies in the OFF state are needed to further investigate the broader impact of cross‐talk on FOG expression.

A ceiling effect of task could also help to explain our results. The auditory Stroop task used to enhance cognitive load imposes substantial executive demands (Kestens et al. 2021), which may have diverted attention away from the TOS when both tasks were administered simultaneously. Cognitive and limbic load interact closely, even in healthy people (Norouzian et al. 2024; Young and Williams 2015) and anxiety can either promote cautious behavior or bias attention towards threatening, task‐irrelevant stimuli (Eysenck et al. 2007; Young and Williams 2015). This complex interplay between attention and anxiety is likely more pronounced in people with PD and FOG, given their impaired automatic gait control and reduced cognitive capacity (Peterson et al. 2016; Vandenbossche et al. 2012; Wu et al. 2015).

During the 360° turns, TOS significantly elevated the FOGratio, whereas the DT did not. This may reflect the greater variability in DT performance, driven by individual differences in cognitive ability (Monaghan et al. 2023) and poorer cognition in freezers. Variability may also have been amplified by the 360° turns themselves, known to be highly FOG provoking and heightening worrisome thoughts and fear (Eysenck et al. 2007; Young and Williams 2015). Another possibility is that the 360° turns were already cognitively demanding, leaving limited scope for the DT to add further load.

A somewhat counterintuitive finding was that the TOS tended to worsen the FOGratio, yet participants completed the FOG‐protocol faster. We interpret this as participants attempting to rush through the threat conditions to reduce shock‐induced anxiety, a pattern also reported in healthy subjects (Robinson et al. 2013). Although fear can sometimes slow gait velocity, such as when navigating precision‐demanding environments with a high risk of falling, like walking on a high plank (Norouzian et al. 2024), the TOS paradigm used here likely provoked a less intense or qualitatively different form of anxiety. Specifically, it appeared to increase urgency without altering gait mechanics, as it offered a “cleaner” manipulation of anxiety that did not interfere directly with the walking pattern.

Contrary to our expectations, the state anxiety induced by our threat paradigms did not differ between groups. Trait anxiety, as assessed using PAS scores, was likewise comparable between freezers and nonfreezers, while other nonmotor features, such as cognition and autonomic function, were significantly more impaired in freezers, consistent with the literature (Ehgoetz Martens et al. 2020). In previous study, higher trait anxiety has been linked to more cautious gait during fear‐inducing tasks (Norouzian et al. 2024). In our study, however, adding the PAS score (and its interactions) to the statistical model did not improve model fit, suggesting that trait anxiety did not modulate the FOGratio. This may reflect that trait anxiety is more closely related to fear of falling than to a more generic threat. A recent systematic review similarly noted substantial variability in trait anxiety findings in freezers (Witt et al. 2019), likely due to differences in study populations or in the primary outcome measures.

Previous work reported increased self‐reported anxiety in freezers compared with nonfreezers (Ehgoetz Martens et al. 2014), and it has been suggested that the higher falls risk experienced by freezers would heighten anxiety during FOG‐provoking situations (Duysens and Smits‐Engelsman 2023). Conversely, we found that freezers' self‐perceived stress was higher only when compared with age‐matched healthy controls and that anxiety and cognitive load produced similar effects across groups. This contrasts with recent suggestions that the limbic system in freezers becomes hypersensitive through repeated exposure to FOG‐triggering situations in daily life (Duysens and Smits‐Engelsman 2023; Gilat et al. 2018).

Interestingly, results were not consistent across objective and subjective stress outcomes. Freezers reported higher perceived stress during the protocol, yet their physiological responses were comparable with those of the other groups. Moreover, while GSR showed a similar increase in stress for the DT, TOS, and DT + TOS relative to baseline, patients perceived the combined DT + TOS as even more stressful. In contrast, during the threat of the “FOG‐eliciting braces” (TOF), patients reported stress levels similar to baseline, despite increases in GSR. HR was also elevated only during TOF and not during TOS or DT.

There are several potential reasons for these inconsistencies. First, the order of conditions was only partly counterbalanced, and the higher physiological response during TOF may have reflected cumulative effort rather than the threat itself. Second, the threat posed by TOF may have been insufficiently convincing, possibly because participants did not fully believe that the braces would actually induce FOG. However, because no formal, validated measure of perceived threat credibility was administered, we cannot confirm this explanation. Another possibility is that participants rated perceived stress as lower during TOF because the threat was never enacted in the trials of interest, unlike during the TOS. The braces were only activated during the two initial trials designed to establish the “threat of FOG.” Third, it should be noted that the physiological stress sensors have so far been validated in healthy subjects only and not in people with PD (Smets et al. 2019). Furthermore, the short computational window (i.e., task duration) may have influenced discrepant results for HR and GSR, as HR is based on the number of peaks per window and is therefore more sensitive to window duration. Future studies are needed to better interpret the physiological outcomes from wearable sensors, in particular, in relation to PD‐specific factors including medication effects and autonomic dysfunction.

We found that the TOF did not negatively affect FOG‐related outcomes in contrast to TOS (when compared with baseline). Particularly, %TF was significantly better in the TOF than in the TOS. Three nonmutually exclusive explanations may account for this result. First, as noted above, the threat of the TOF may have been inadequate, also supported by the observed lower perceived stress levels, though not by physiological markers. Second, according to the well‐established inverted U‐shaped relationship between arousal and motor performance, TOF may have induced a more optimal arousal level than TOS (Tosserams et al. 2023), improving motor performance despite the presence of stress. Finally, TOF may have elicited a different type of response that protected freezers against FOG. This interpretation aligns with the attentional control theory, which proposes that performance is shaped by the attentional bias towards threat (Eysenck et al. 2007). A gait‐related threat may therefore have been more beneficial than a nongait‐related one, as it directed attention towards walking. In addition, freezers' familiarity with freezing may have enabled them to compensate more effectively, reducing %TF, whereas in the other groups, TOF may have generated primarily worry and distraction.

Several limitations warrant cautious interpretation of the results. First, we used a higher frequency of exposure to shocks than in previous work (Robinson et al. 2013; Schmitz and Grillon 2012; van Langen 2019) to ensure that the TOS was sufficiently intense, which may have induced some habituation. Second, measurements were performed ON medication to enhance feasibility, but this may have reduced the responsiveness of the FOG outcomes, and it is unclear whether the current results would generalize to the OFF state (Snijders et al. 2012; Zoetewei et al. 2024). Because this was the first study to apply a TOS paradigm to investigate FOG, the sample size could only be determined pragmatically. Finally, the FOGratio may have been influenced by gait speed, FOG manifestation (trembling or akinetic), and measurement noise. Furthermore, its validity as a measure of FOG‐related motor breakdown in nonfreezer groups warrants further investigation. Nonetheless, FOGratios were higher in freezers than in the other groups and were higher during 360° turns (D'Cruz et al. 2022; van Dijsseldonk et al. 2018), supporting its validity and use as the primary outcome.

5. Conclusion

TOS and cognitive load both tended to worsen the FOGratio, but their combination did not enhance these effects, which is inconsistent with the cross‐talk hypothesis (Shine et al. 2011). State anxiety may affect FOG via attentional control rather than overload, or the current paradigm may not have generated enough load to meaningfully disrupt gait (Eysenck et al. 2007; Tosserams et al. 2023). Contrary to expectations, freezers were not more sensitive to moderate levels of anxiety or cognitive load in this study. Further research is needed to understand how anxiety shapes FOG during daily life medication fluctuations, how different triggers interact, and whether treatment of anxiety can improve FOG (van der Heide et al. 2021; Vissers et al. 2026).

Author Contributions

Demi Zoetewei: methodology, software, investigation, formal analysis, writing – original draft, conceptualization, data curation. Moran Gilat: conceptualization, supervision, writing – review and editing, methodology. Nicholas D'Cruz: conceptualization, methodology, writing – review and editing. Jan Cornelis: formal analysis, writing – review and editing. Jordan Bedime Songue Njiki: investigation, formal analysis, writing – review and editing. Rick C. Helmich: writing – review and editing. Pieter Ginis: conceptualization, methodology, supervision, writing – review and editing. Alice Nieuwboer: conceptualization, methodology, supervision, writing – review and editing.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

EJN-64-0-s001.docx (227.5KB, docx)

Video S1: Supporting Information.

Download video file (20.1MB, mov)

Acknowledgments

We would like to express our gratitude towards all the participants involved in the study and particularly the gentleman that acted in Video S1.

Zoetewei, D. , Gilat M., D'Cruz N., et al. 2026. “The Effects of Anxiety and Cognitive Load on Freezing of Gait in Parkinson's Disease.” European Journal of Neuroscience 64, no. 5: e70678. 10.1111/ejn.70678.

Associate Editor: Stewart A. Alan Factor

Data Availability Statement

The data are sharable under restricted access upon reasonable request pending ethical approval and a data sharing agreement in accordance with the European General Data Protection Regulation (GDPR).

References

  1. Beck, A. T. , Ward C. H., Mendelson M., Mock J., and Erbaugh J.. 1961. “An Inventory for Measuring Depression.” Archives of General Psychiatry 4: 561–571. 10.1001/archpsyc.1961.01710120031004. [DOI] [PubMed] [Google Scholar]
  2. Bloem, B. R. , Hausdorff J. M., Visser J. E., and Giladi N.. 2004. “Falls and Freezing of Gait in Parkinson's Disease: A Review of Two Interconnected, Episodic Phenomena.” Movement Disorders 19, no. 8: 871–884. 10.1002/mds.20115. [DOI] [PubMed] [Google Scholar]
  3. Cockx, H. , Nonnekes J., Bloem B. R., van Wezel R., Cameron I., and Wang Y.. 2023. “Dealing With the Heterogeneous Presentations of Freezing of Gait: How Reliable Are the Freezing Index and Heart Rate for Freezing Detection?” Journal of Neuroengineering and Rehabilitation 20, no. 1: 53. 10.1186/s12984-023-01175-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. D'Cruz, N. , Seuthe J., De Somer C., et al. 2022. “Dual Task Turning in Place: A Reliable, Valid, and Responsive Outcome Measure of Freezing of Gait.” Movement Disorders 37, no. 2: 269–278. 10.1002/mds.28887. [DOI] [PubMed] [Google Scholar]
  5. Duysens, J. , and Smits‐Engelsman B.. 2023. “Freezing as Seen From the Inside.” Movement Disorders 38, no. 9: 1598–1601. 10.1002/mds.29444. [DOI] [PubMed] [Google Scholar]
  6. Economou, K. , Quek D., MacDougall H., Lewis S. J. G., and Ehgoetz Martens K. A.. 2021. “Heart Rate Changes Prior to Freezing of Gait Episodes Are Related to Anxiety.” Journal of Parkinson's Disease 11, no. 1: 271–282. 10.3233/JPD-202146. [DOI] [PubMed] [Google Scholar]
  7. Ehgoetz Martens, K. A. , Ellard C. G., and Almeida Q. J.. 2014. “Does Anxiety Cause Freezing of Gait in Parkinson's Disease?” PLoS ONE 9, no. 9: e106561. 10.1371/journal.pone.0106561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ehgoetz Martens, K. A. , Ellard C. G., and Almeida Q. J.. 2015. “Anxiety‐Provoked Gait Changes Are Selectively Dopa‐Responsive in Parkinson's Disease.” European Journal of Neuroscience 42, no. 4: 2028–2035. 10.1111/ejn.12928. [DOI] [PubMed] [Google Scholar]
  9. Ehgoetz Martens, K. A. , Lukasik E. L., Georgiades M. J., et al. 2018. “Predicting the Onset of Freezing of Gait: A Longitudinal Study.” Movement Disorders 33, no. 1: 128–135. 10.1002/mds.27208. [DOI] [PubMed] [Google Scholar]
  10. Ehgoetz Martens, K. A. , Peterson D. S., Almeida Q. J., Lewis S. J. G., Hausdorff J. M., and Nieuwboer A.. 2020. “Behavioural Manifestations and Associated Non‐Motor Features of Freezing of Gait: A Narrative Review and Theoretical Framework.” Neuroscience and Biobehavioral Reviews 116: 350–364. 10.1016/j.neubiorev.2020.06.026. [DOI] [PubMed] [Google Scholar]
  11. ELAN . 2019. Version 5.5. Max Planck Institute for Psycholinguistics, The Language Archive. [Google Scholar]
  12. El‐Gohary, M. , Pearson S., McNames J., et al. 2014. “Continuous Monitoring of Turning in Patients With Movement Disability.” Sensors 14, no. 1: 356–369. 10.3390/s140100356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eysenck, M. W. , Derakshan N., Santos R., and Calvo M. G.. 2007. “Anxiety and Cognitive Performance: Attentional Control Theory.” Emotion (Washington, D.C.) 7, no. 2: 336–353. 10.1037/1528-3542.7.2.336. [DOI] [PubMed] [Google Scholar]
  14. Folstein, M. F. , Folstein S. E., and McHugh P. R.. 1975. ““Mini‐Mental State”: A Practical Method for Grading the Cognitive State of Patients for the Clinician.” Journal of Psychiatric Research 12, no. 3: 189–198. 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  15. Gilat, M. 2019. “How to Annotate Freezing of Gait From Video: A Standardized Method Using Open‐Source Software.” Journal of Parkinson's Disease 9, no. 4: 821–824. 10.3233/JPD-191700. [DOI] [PubMed] [Google Scholar]
  16. Gilat, M. , Ehgoetz Martens K. A., Miranda‐Domínguez O., et al. 2018. “Dysfunctional Limbic Circuitry Underlying Freezing of Gait in Parkinson's Disease.” Neuroscience 374: 119–132. 10.1016/j.neuroscience.2018.01.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gilat, M. , Nonnekes J., Factor S. A., et al. 2026. “An Updated Definition of Freezing of Gait.” Nature Reviews Neurology 22: 172–181. 10.1038/s41582-025-01179-3. [DOI] [PubMed] [Google Scholar]
  18. Goetz, C. G. , Tilley B. C., Shaftman S. R., et al. 2008. “Movement Disorder Society‐Sponsored Revision of the Unified Parkinson's Disease Rating Scale (MDS‐UPDRS): Scale Presentation and Clinimetric Testing Results.” Movement Disorders 23, no. 15: 2129–2170. 10.1002/mds.22340. [DOI] [PubMed] [Google Scholar]
  19. Hoehn, M. M. , and Yahr M. D.. 1967. “Parkinsonism: Onset, Progression, and Mortality.” Neurology 17, no. 5: 427–442. 10.1212/WNL.17.5.427. [DOI] [PubMed] [Google Scholar]
  20. Hughes, A. J. , Daniel S. E., Kilford L., and Lees A. J.. 1992. “Accuracy of Clinical Diagnosis of Idiopathic Parkinson's Disease: A Clinico‐Pathological Study of 100 Cases.” Journal of Neurology Neurosurgery and Psychiatry 55, no. 3: 181–184. 10.1136/jnnp.55.3.181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jankovic, J. , McDermott M., Carter J., et al. 1990. “Variable Expression of Parkinson's Disease: A Base‐Line Analysis of the DATATOP Cohort. The Parkinson Study Group.” Neurology 40, no. 10: 1529–1534. 10.1212/wnl.40.10.1529. [DOI] [PubMed] [Google Scholar]
  22. Jost, S. T. , Kaldenbach M. A., Antonini A., et al. 2023. “Levodopa Dose Equivalency in Parkinson's Disease: Updated Systematic Review and Proposals.” Movement Disorders 38, no. 7: 1236–1252. 10.1002/mds.29410. [DOI] [PubMed] [Google Scholar]
  23. Kestens, K. , Degeest S., Miatton M., and Keppler H.. 2021. “An Auditory Stroop Test to Implement in Cognitive Hearing Sciences: Development and Normative Data.” International Journal of Psychological Research 14, no. 2: 37–51. 10.21500/20112084.5118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Klumpers, F. , Raemaekers M. A. H. L., Ruigrok A. N. V., Hermans E. J., Kenemans J. L., and Baas J. M. P.. 2010. “Prefrontal Mechanisms of Fear Reduction After Threat Offset.” Biological Psychiatry 68, no. 11: 1031–1038. 10.1016/j.biopsych.2010.09.006. [DOI] [PubMed] [Google Scholar]
  25. Lakens, D. 2013. “Calculating and Reporting Effect Sizes to Facilitate Cumulative Science: A Practical Primer for t‐Tests and ANOVAs.” Frontiers in Psychology 4: 863. 10.3389/fpsyg.2013.00863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Leentjens, A. F. G. , Dujardin K., Pontone G. M., Starkstein S. E., Weintraub D., and Martinez‐Martin P.. 2014. “The Parkinson Anxiety Scale (PAS): Development and Validation of a New Anxiety Scale.” Movement Disorders 29, no. 8: 1035–1043. 10.1002/mds.25919. [DOI] [PubMed] [Google Scholar]
  27. Lewis, S. J. G. , and Shine J. M.. 2016. “The Next Step: A Common Neural Mechanism for Freezing of Gait.” Neuroscientist 22, no. 1: 72–82. 10.1177/1073858414559101. [DOI] [PubMed] [Google Scholar]
  28. Maidan, I. , Plotnik M., Mirelman A., Weiss A., Giladi N., and Hausdorff J. M.. 2010. “Heart Rate Changes During Freezing of Gait in Patients With Parkinson's Disease.” Movement Disorders 25, no. 14: 2346–2354. 10.1002/mds.23280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mancini, M. , Hasegawa N., Peterson D. S., Horak F. B., and Nutt J. G.. 2023. “Digital Measures of Freezing of Gait Across the Spectrum of Normal, Non‐Freezers, Possible Freezers and Definite Freezers.” Journal of Neurology 270, no. 9: 4309–4317. 10.1007/s00415-023-11773-4. [DOI] [PubMed] [Google Scholar]
  30. Mancini, M. , Priest K. C., Nutt J. G., and Horak F. B.. 2012. “Quantifying Freezing of Gait in Parkinson's Disease During the Instrumented Timed Up and Go Test.” In 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1198–1201. IEEE. 10.1109/EMBC.2012.6346151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mancini, M. , Smulders K., Cohen R. G., Horak F. B., Giladi N., and Nutt J. G.. 2017. “The Clinical Significance of Freezing While Turning in Parkinson's Disease.” Neuroscience 343: 222–228. 10.1016/j.neuroscience.2016.11.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mazilu, S. , Calatroni A., Gazit E., Mirelman A., Hausdorff J. M., and Tröster G.. 2015. “Prediction of Freezing of Gait in Parkinson's From Physiological Wearables: An Exploratory Study.” IEEE Journal of Biomedical and Health Informatics 19, no. 6: 1843–1854. 10.1109/JBHI.2015.2465134. [DOI] [PubMed] [Google Scholar]
  33. McDonald, L. M. , Griffin H. J., Angeli A., Torkamani M., Georgiev D., and Jahanshahi M.. 2015. “Motivational Modulation of Self‐Initiated and Externally Triggered Movement Speed Induced by Threat of Shock: Experimental Evidence for Paradoxical Kinesis in Parkinson's Disease.” PLoS ONE 10, no. 8: e0135149. 10.1371/journal.pone.0135149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Monaghan, A. S. , Ragothaman A., Harker G. R., Carlson‐Kuhta P., Horak F. B., and Peterson D. S.. 2023. “Freezing of Gait in Parkinson's Disease: Implications for Dual‐Task Walking.” Journal of Parkinson's Disease 13, no. 6: 1035–1046. 10.3233/JPD-230063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nieuwboer, A. , Rochester L., Herman T., et al. 2009. “Reliability of the New Freezing of Gait Questionnaire: Agreement Between Patients With Parkinson's Disease and Their Carers.” Gait and Posture 30, no. 4: 459–463. 10.1016/j.gaitpost.2009.07.108. [DOI] [PubMed] [Google Scholar]
  36. Nieuwhof, F. , Bloem B. R., Reelick M. F., et al. 2017. “Impaired Dual Tasking in Parkinson's Disease Is Associated With Reduced Focusing of Cortico‐Striatal Activity.” Brain 140, no. 5: 1384–1398. 10.1093/brain/awx042. [DOI] [PubMed] [Google Scholar]
  37. Norouzian, P. , Horslen B. C., and Martens K. A. E.. 2024. “The Effects of Trait and State Anxiety on Gait in Healthy Young Adults.” Experimental Brain Research 242, no. 4: 819–828. 10.1007/s00221-024-06800-3. [DOI] [PubMed] [Google Scholar]
  38. Peterson, D. S. , King L. A., Cohen R. G., and Horak F. B.. 2016. “Cognitive Contributions to Freezing of Gait in Parkinson Disease: Implications for Physical Rehabilitation.” Physical Therapy 96, no. 5: 659–670. 10.2522/ptj.20140603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Robinson, O. J. , Vytal K., Cornwell B. R., and Grillon C.. 2013. “The Impact of Anxiety Upon Cognition: Perspectives From Human Threat of Shock Studies.” Frontiers in Human Neuroscience 7: 203. 10.3389/fnhum.2013.00203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schmitz, A. , and Grillon C.. 2012. “Assessing Fear and Anxiety in Humans Using the Threat of Predictable and Unpredictable Aversive Events (the NPU‐Threat Test).” Nature Protocols 7, no. 3: 527–532. 10.1038/nprot.2012.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shine, J. M. , Naismith S. L., and Lewis S. J. G.. 2011. “The Pathophysiological Mechanisms Underlying Freezing of Gait in Parkinson's Disease.” Journal of Clinical Neuroscience 18, no. 9: 1154–1157. 10.1016/j.jocn.2011.02.007. [DOI] [PubMed] [Google Scholar]
  42. Smets, E. , De Raedt W., and Van Hoof C.. 2019. “Into the Wild: The Challenges of Physiological Stress Detection in Laboratory and Ambulatory Settings.” IEEE Journal of Biomedical and Health Informatics 23, no. 2: 463–473. 10.1109/JBHI.2018.2883751. [DOI] [PubMed] [Google Scholar]
  43. Snijders, A. H. , Haaxma C. A., Hagen Y. J., Munneke M., and Bloem B. R.. 2012. “Freezer or Non‐Freezer: Clinical Assessment of Freezing of Gait.” Parkinsonism and Related Disorders 18, no. 2: 149–154. 10.1016/j.parkreldis.2011.09.006. [DOI] [PubMed] [Google Scholar]
  44. Taylor, N. L. , Wainstein G., Quek D., Lewis S. J. G., Shine J. M., and Ehgoetz Martens K. A.. 2022. “The Contribution of Noradrenergic Activity to Anxiety‐Induced Freezing of Gait.” Movement Disorders 37, no. 7: 1432–1443. 10.1002/mds.28999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Tosserams, A. , Bloem B. R., Ehgoetz Martens K. A., et al. 2023. “Modulating Arousal to Overcome Gait Impairments in Parkinson's Disease: How the Noradrenergic System May Act as a Double‐Edged Sword.” Translational Neurodegeneration 12, no. 1: 15. 10.1186/s40035-023-00347-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. van der Heide, A. , Speckens A. E. M., Meinders M. J., Rosenthal L. S., Bloem B. R., and Helmich R. C.. 2021. “Stress and Mindfulness in Parkinson's Disease—A Survey in 5000 Patients.” npj Parkinson's Disease 7, no. 1: 7. 10.1038/s41531-020-00152-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. van Dijsseldonk, K. , Wang Y., van Wezel R., Bloem B. R., and Nonnekes J.. 2018. “Provoking Freezing of Gait in Clinical Practice: Turning in Place Is More Effective Than Stepping in Place.” Journal of Parkinson's Disease 8, no. 2: 363–365. 10.3233/JPD-181332. [DOI] [PubMed] [Google Scholar]
  48. van Langen, J. 2019. The Influence of Stress on Parkinson's Tremor, Master Thesis. Radboud University. 10.13140/RG.2.2.33608.96001. [DOI] [Google Scholar]
  49. Vandenbossche, J. , Deroost N., Soetens E., et al. 2012. “Freezing of Gait in Parkinson's Disease: Disturbances in Automaticity and Control.” Frontiers in Human Neuroscience 6: 356. 10.3389/fnhum.2012.00356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Visser, M. , Marinus J., Stiggelbout A. M., and van Hilten J. J.. 2004. “Assessment of Autonomic Dysfunction in Parkinson's Disease: The SCOPA‐AUT.” Movement Disorders 19, no. 11: 1306–1312. 10.1002/mds.20153. [DOI] [PubMed] [Google Scholar]
  51. Vissers, G. , Tosserams A., Duits A. A., et al. 2026. “Tackling Anxiety‐ and Stress‐Related Freezing of Gait in People With Parkinson's Disease (TACKLING‐FOG): Study Protocol for a Randomized Controlled Trial.” European Journal of Neuroscience 63, no. 12: e70584. 10.22541/au.176866755.56870836/v1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Witt, I. , Ganjavi H., and MacDonald P.. 2019. “Relationship Between Freezing of Gait and Anxiety in Parkinson's Disease Patients: A Systemic Literature Review.” Parkinson's Disease 2019, no. 1: 1–24. 10.1155/2019/6836082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wu, T. , Hallett M., and Chan P.. 2015. “Motor Automaticity in Parkinson's Disease.” Neurobiology of Disease 82: 226–234. 10.1016/j.nbd.2015.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Young, W. R. , and Williams A. M.. 2015. “How Fear of Falling Can Increase Fall‐Risk in Older Adults: Applying Psychological Theory to Practical Observations.” Gait and Posture 41, no. 1: 7–12. 10.1016/j.gaitpost.2014.09.006. [DOI] [PubMed] [Google Scholar]
  55. Ziegler, K. , Schroeteler F., Ceballos‐Baumann A. O., and Fietzek U. M.. 2010. “A New Rating Instrument to Assess Festination and Freezing Gait in Parkinsonian Patients.” Movement Disorders 25, no. 8: 1012–1018. 10.1002/mds.22993. [DOI] [PubMed] [Google Scholar]
  56. Zoetewei, D. , Ginis P., Goris M., et al. 2024. “Which Gait Tasks Produce Reliable Outcome Measures of Freezing of Gait in Parkinson's Disease?” Journal of Parkinson's Disease 14, no. 6: 1163–1174. 10.3233/JPD-240134. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: Supporting Information.

EJN-64-0-s001.docx (227.5KB, docx)

Video S1: Supporting Information.

Download video file (20.1MB, mov)

Data Availability Statement

The data are sharable under restricted access upon reasonable request pending ethical approval and a data sharing agreement in accordance with the European General Data Protection Regulation (GDPR).

Articles from The European Journal of Neuroscience are provided here courtesy of Wiley

RESOURCES