Neuromodulation is a type of intervention that relies on various non-invasive techniques to temporarily stimulate the brain and nervous system. It can be used for the treatment of depression or other medical conditions, as well as the improvement of cognitive abilities such as attention. However, there is conflicting evidence regarding whether this approach has beneficial effects.

Most studies aiming to assess the efficiency of a treatment rely on examining the outcomes of people who received the intervention in comparison to participants who undergo a similar procedure with no therapeutic effect (or placebo). However, the influence of other, ‘subjective’ factors on these results – such as the type of intervention participants think they have received – remains poorly investigated.

To bridge this gap, Fassi and Hochman et al. used statistical modeling to assess how patients’ beliefs about their treatment affected the results of four neuromodulation studies on mind wandering, depression and attention deficit hyperactivity disorder symptoms. In two studies, participants' perceptions of their treatment status were more strongly linked to changes in depression scores and mind-wandering than the actual treatment. Results were more nuanced in the other two studies. In one of them, participants who received the real neuromodulation but believed they received the placebo showed the most improvement in depressive symptoms; in the other study, subjective beliefs and objective treatment both explained changes in inattention symptoms.

Taken together, the results by Fassi and Hochman et al. suggest that factoring in patients’ subjective beliefs about their treatment may be necessary in studies of neuromodulation and other interventions like virtual reality or neurofeedback, where participants are immersed in cutting-edge research settings and might therefore be more susceptible to develop beliefs about treatment efficacy.

A substantial amount of research from medicine, neuroscience, psychology, and education aims to establish the effectiveness of different treatments, such as drugs, cognitive training, biofeedback, and neurostimulation, in both clinical and non-clinical populations. However, the research findings from these fields tend to be heterogeneous. As a result, there has been increased scepticism among researchers about the efficacy of these treatments (Lampit et al., 2014; López-Alonso et al., 2014; Sitaram et al., 2017).

In recent years, neuromodulation has been studied as one of the most promising treatment methods (De Ridder et al., 2021). Further, one particular form of neuromodulation, transcranial magnetic stimulation (TMS), has been approved by regulatory bodies in multiple countries, including the US Food and Drug Administration (FDA), and is used as an evidence-based treatment for patients with migraine, major depression, obsessive-compulsive disorder, and smoking addiction (Hallett, 2007; Walsh and Cowey, 2000). Moreover, TMS and other neuromodulatory devices, such as transcranial-focused ultrasound and electrical stimulation (tES), have been highlighted as a potential treatment for psychiatric, neurological, and neurodevelopmental disorders (Grover et al., 2021; Khedr et al., 2005; McGough et al., 2019), and they have also been used to enhance various mental processes, including attention, memory, language, mathematics, and intelligence in healthy populations (Santarnecchi et al., 2015). These encouraging findings have raised hope for the potential application of these techniques within and outside the clinic (Dubljević et al., 2014).

Despite some encouraging results on the beneficial effects of both TMS and tES, contradictory findings have emerged across different studies (Horvath et al., 2015; Medina and Cason, 2017; Parkin et al., 2015; Wang et al., 2018; Westwood et al., 2017). Several factors have been pointed at as plausible reasons for the heterogeneity in research results (Filmer et al., 2020; Guerra et al., 2020; van Bueren et al., 2021). However, a crucial factor that researchers have largely overlooked is the extent to which subjective beliefs can explain variability in treatment efficacy. Here, we address this gap by examining whether modelling participants’ beliefs about receiving the placebo or active treatment can account for changes in clinical, cognitive and behavioural outcomes.

Participants who take part in TMS and tES studies consistently report various perceptual sensations, such as audible clicks, visual disturbances, and cutaneous sensations (Davis et al., 2013). Consequently, they can discern when they have received the active treatment, making subjective beliefs and demand characteristics potentially influencing performance (Polanía et al., 2018). To account for such non-specific effects, sham (placebo) protocols have been employed. For transcranial direct current stimulation (tDCS), the most common form of tES, various sham protocols exist. A review by Fonteneau et al., 2019 shows that 84% of 173 studies used similar sham approaches to an early method by Gandiga et al., 2006. This initial protocol had a 10 s ramp-up followed by 30 s of active stimulation at 1 mA before cessation, differently from active stimulation that typically lasts up to 20 min. However, this has been adapted in terms of intensity and duration of current, ramp-in/-out phases, and the number of ramps during stimulation. Similarly, in sham TMS, the TMS coil may be tilted or replaced with purpose-built sham coils equipped with magnetic shields, which produce auditory effects but ensure no brain stimulation (Duecker and Sack, 2015). By using surface electrodes, the somatosensory effects of actual TMS can also be mimicked. Overall, these types of sham stimulation aim to simulate the perceptual sensations associated with active stimulation without substantially affecting cortical excitability (Fritsch et al., 2010; Nitsche and Paulus, 2000). As a result, sham treatments should allow controlling for participants’ specific beliefs about the type of stimulation received.

Previous studies have addressed whether manipulating participants’ expectations about the effects of either active or sham stimulation can moderate treatment efficacy (Haikalis et al., 2023; Rabipour et al., 2018). However, to our knowledge, these studies have not examined whether individual differences in participants’ subjective experience of receiving the active or sham treatment provide a better model fit than the condition to which participants are assigned in the study. We term the former subjective treatment and the latter objective treatment.

The above consideration becomes particularly crucial when considering that the experimental design of most randomised controlled trials (RCTs) involves recording whether participants believed they received the active or placebo treatment. While it is common practice to assess experimental blinding using this data, the explanatory power of individual differences in subjective treatment is rarely, if at all, considered. This is based on the assumption that if no differences emerged at the group level in participants’ guess for receiving the active vs the placebo treatment (i.e. if experimental blinding was successful), placebo effects could not explain the obtained results.

Here, we hypothesise that such an assumption can be erroneous and aim to explore how accounting for differences in subjective beliefs can shed light on the conclusions of previous treatment studies. Moreover, we introduce a simple and straightforward approach that could be used to analyse existing data and guide future clinical and fundamental research to examine whether subjective treatment explains variability in experimental outcomes over and beyond objective treatment. Below, we demonstrate this approach by reanalysing four independently published neurostimulation studies (including TMS and tES) that test clinical and non-clinical samples from different age groups (Blumberger et al., 2016; Filmer et al., 2019; Leffa et al., 2022; Kaster et al., 2018). The data and the codebook of the analyses are available on the OSF (https://osf.io/rztxu/).

In this work, we used a novel approach to examine whether and to what extent participants’ subjective beliefs may account for variability in research outcomes. To this aim, we analysed four independent datasets from the field of neurostimulation; specifically, two rTMS RCTs in patients with depression (Blumberger et al., 2016; Kaster et al., 2018), one tDCS study in adults with ADHD (Leffa et al., 2022), and another tDCS study in a healthy adult sample (Filmer et al., 2019).

We demonstrate that participants’ subjective beliefs about receiving the active vs control (sham) treatment are an important factor that can explain variability in the primary outcome and, in some cases, fit the observed data better than the actual treatment participants received during the experiment. Specifically, in studies 1 and 4, the fact that participants thought to be in the active or control condition explained variability in clinical and cognitive scores to a more considerable extent than the objective treatment alone. Notably, the same pattern of results emerged when we replaced subjective treatment with subjective dosage in the fourth experiment, showing that subjective beliefs about treatment intensity also explained variability in research results better than objective treatment. In contrast to studies 1 and 4, studies 2 and 3 showed a more complex pattern of results. Specifically, in study 2 we observed an interaction effect, whereby the greatest improvement in depressive symptoms was observed in the group that received the active objective treatment but believed they received sham. Differently, in study 3, the inclusion of both subjective and objective treatment as main effects explained variability in symptoms of inattention. Overall, these findings suggest the complex interplay of objective and subjective treatment. The variability in the observed results could be explained by factors such as participants’ personality, disorder type and severity, prior treatments, knowledge base, experimental procedures, and beliefs of the research team, all of which could be interesting avenues for future studies to explore.

An important question arising from our findings relates to the causal role of subjective beliefs. This question is a complex one to answer and falls outside the scope of this study. Based on the goal of testing blinding efficacy, it is a standard practice for current treatment studies to record data on subjective beliefs only at the end of the experiment rather than before and throughout. This was the case in studies 1, 3, and 4. While in study 3, both objective and subjective treatment explained unique variance in the clinical outcomes, in studies 1 and 4 it was impossible to conclude whether participants’ beliefs, which were captured by the subjective treatment, affected experimental outcomes or, on the contrary, whether participants’ changes in performance or symptoms throughout the studies influenced beliefs regarding treatment allocation. Therefore, it is important to consider that subjective treatment could also capture the changes faced by participants in placebo-controlled trials, whereby, if they feel better, it would be hard psychologically to report that they believe it was due to the placebo. Notably, study 2 used a less common approach, in which subjective beliefs about treatment allocation were queried after the first week of the study. In this case, significant results emerged only 2 weeks after this inquiry (i.e., weeks 3 and 4). Given that subjective beliefs about treatment allocation were documented before the emergence of changes in any clinical outcome, it is more plausible that participants’ beliefs would have affected experimental outcomes rather than vice versa.

Based on the above considerations, future studies should strive to record data on subjective beliefs at different time points: before, during, and after the experiment. This will allow mapping the way subjective beliefs might be differentially associated with experimental results depending on the considered study and treatment type. However, we acknowledge two caveats of this suggestion. Firstly, participants may be more prone to pay attention to their treatment allocation and, consequently, figure out their assigned condition. Secondly, recording subjective beliefs at multiple time points might interfere with the effects of the treatment. For instance, patients might suppress their response for the fear that the treatment received is a placebo (Sonawalla and Rosenbaum, 2002). An alternative approach could entail deception, whereby all participants are told they received the active treatment. While this raises an ethical concern, such an approach would: (1) allow minimising the effect of subjective beliefs on research outcomes, and (2) hold more ecological validity, as it would mimic the way approved treatments are delivered in the clinic, where all patients know to be receiving the active treatment (Fonteneau et al., 2019).

While this study focuses only on neuromodulation techniques, we want to highlight that the proposed approach can be applied to other forms of treatment (e.g. pharmacological studies, cognitive training) tested as part of standard experiments or RCTs. It is worth noting that the contribution of subjective beliefs to experimental results might be even more enhanced when considering interventions carried out in seemly cutting-edge research settings, such as experiments involving virtual reality, neurofeedback paradigms, and other types of brain–computer interfaces. In such cases, participants might be more susceptible to forming specific expectations about treatment effects (Fonteneau et al., 2019; Thibault et al., 2017; Mikellides et al., 2022). Therefore, the explanatory power of subjective beliefs could be intensified compared to more traditional forms of treatment, such as pharmacology.

One question emerging from this study is whether these results, observed with self-report measures, would apply to more objective behavioural outcomes (e.g. sensorimotor recovery in stroke patients, improvements in fluid intelligence in participants with learning difficulties) and neural functions (e.g. functional connectivity). We argue that the answer to this question is likely to be positive since placebo effects have been shown to impact not only behaviour but also brain activity (Fonteneau et al., 2019; Hashmi, 2018; Oken et al., 2008; Schmidt et al., 2014). However, independent of this possibility, the contribution of subjective treatment to explaining variability in self-reported outcomes should not be underestimated. Noteworthy, in most RCTs investigating the effect of different treatments on clinical and subclinical groups (e.g. depression, chronic pain, eating disorders, attention-deficit/hyperactivity disorder), some of which have also been approved by the FDA, the measurement of symptomatology is mostly based on self-reported outcomes, such as questionnaires. This consideration makes the case of subjective treatment even stronger, hinting at the potential role of this factor in explaining experimental results across a variety of experimental outcomes and treatment types.

While our study examined the explanatory power of subjective beliefs about receiving treatment, neither of the four studies (similar to most studies in the field) collected data on participants’ expectations. Indeed, as recently shown by Parong et al., 2022, expectations regarding the effectiveness of cognitive training (i.e. whether it will increase or decrease performance) can significantly modulate its effect. Thus, investigating the interplay between expectations and subjective treatment could allow examining the directionality and strength of the effect of subjective treatment on the outcome of interest. For instance, some participants may expect a treatment to improve their capabilities or symptoms. In contrast, others could expect even the opposite, and the level of these expectations can vary during the intervention. These factors could, in turn, impact individual variability in subjective treatment. Arguably, when questioned early, subjective treatment could be more related to expectations rather than an actual reflection of the treatment benefits. This variation may explain the findings in study 2 (decrements in depression for subjective sham treatment) compared to study 1 (decrements in depression for subjective active treatment), where only in the former were subjects questioned during the procedure (week 1) and not at its end. This possibility is a post hoc explanation, and future experiments collecting data on participants’ behavioural, cognitive and clinical outcomes should also record subjective expectations thoroughly (Boot et al., 2013Braga et al., 2021).

We want to highlight that, while we present subjective treatment as an important variable with explanatory power in addition to objective treatment, these results do not imply that participants’ subjective beliefs can explain all of the variability in research outcomes (see also in Hochman et al., in preparation; commenting to Gordon et al., 2022). This is demonstrated in study 3, where the objective treatment significantly explained inattention symptoms even when subjective treatment was accounted for. Additionally, we present in Appendix 1 an example of a neuromodulation study in which objective treatment explained variability in treatment effects that could not be attributed to subjective treatment (Murphy et al., 2020). Based on this consideration, where researchers have data, examining variability in participants’ subjective treatment may add further insight into prior results. However, unsurprisingly, when researchers were contacted about providing data on subjective treatment, many reported that the assessment of subjective beliefs, aside from side effects, was not recorded. Indeed, even our group’s procedure in the past lacked the recording of subjective treatment (e.g. Looi and Cohen Kadosh, 2016; Cohen Kadosh et al., 2007).

Overall, our findings hold twofold importance. Firstly, we introduce two new concepts in the academic literature: subjective treatment and subjective dosage. Secondly, we cast light on the role of participants’ subjective experience in explaining the variability of results from RCTs and experiments that test the effectiveness of neuromodulation on mental health and behaviour. Altogether, we call for future studies to systematically collect data on participants’ subjective beliefs and expectations. Studies that have collected data on subjective beliefs at the start and end of the intervention may consider examining the potential contribution of such beliefs to their results. Aside from estimating subjective beliefs about belonging to the active or control condition, we suggest that future research may consider collecting and analysing data on : (1) participants’ beliefs before and at some midpoint during the experiment rather than only at the end, and (2) participants’ expectations about the directionality and strength of the effect of subjective treatment on expected outcomes. This approach would be enhanced with designs that include deception, whereby all participants are told that they received the active treatment. However, such designs require careful ethical review, particularly in clinical populations. Overall, such data will allow a thorough examination of subjective beliefs, yielding more valid and replicable results to progress scientific and clinical studies to benefit human health and behaviour.

We here apply our approach to experimental study that examined the effect of non-invasive brain stimulation (NIBS) on participants’ working memory. In this case, we show that the inclusion of subjective treatment did not explained variability in experimental outcomes.

This study (Murphy et al., 2020) compared the effects of different NIBS techniques, namely anodal tDCS, tRNS + DC-offset, or sham stimulation over the left DLPFC on working memory (WM) performance and task-related EEG oscillatory activity in 49 healthy adults. Participants were allocated to receive either anodal tDCS (N = 16), high-frequency tRNS + DC-offset (N = 16), or sham stimulation (N = 17) to the left DLPFC using a between-subjects design. The Sternberg WM task was used to measure changes in WM performance before, 5 min and 25 min post-stimulation. Moreover, event-related synchronisation/desynchronisation (ERS/ERD) of oscillatory activity was analysed from EEG recorded during WM encoding and maintenance. At the end of the experiment, participants were asked whether they thought they received active or sham stimulation (presented as a binary choice) via a short questionnaire.

We employed the same analytical approach as for experiments 1–4, looking at the contribution of subjective treatment to model fit over and beyond objective treatment and examining whether objective treatment explained variability in experimental outcomes (WM accuracy and reaction times) over and beyond subjective treatment. A linear mixed model with WM reaction time was fitted to the data for time 0–2 (before NIBS, 5 min after, and 25 min after). The baseline model included objective treatment (sham/tRNS/tDCS) as the main effect, as well as the interaction of time and objective treatment. We first added to this model subjective treatment (active or placebo) as a main effect. Next, we extended the model to include the two-way interaction of time and subjective treatment. Lastly, we considered a model with the three-way interaction of time, subjective treatment, and objective treatment. Our results showed that the inclusion of subjective treatment did not lead to a better model fit neither as a main effect (df = 13, BIC = 1824.16, AIC = 1785.28, p = 0.670) nor as a two-way interaction with time (df = 15, BIC = 1833.25, AIC = 1788.40, p = 0.642) and a three-way interaction with time and objective treatment (df = 21, BIC = 1856.48, AIC = 1793.70, p = 0.348). Hence, participants’ subjective experience about the treatment did not explain variability in reaction times beyond the actual treatment condition to which participants were assigned.

We next examined whether the changes in reaction time were explained by objective treatment. We, therefore, added objective treatment first and, following, objective treatment * time after subjective treatment * time was already included in the baseline model. Our results showed that the inclusion of neither objective treatment (df = 11, BIC = 1820.58, AIC = 1787.69, p = 0.139) nor objective treatment * time (df = 15, BIC = 1833.25, AIC = 1788.40, p = 0.121) led to a better model fit. Therefore, when accounting for participants’ subjective experience of receiving the real or placebo treatment, the actual treatment to which participants were assigned during the experiment did not contribute to explaining changes in WM reaction time.

Following, we run the same analysis with WM accuracy as the outcome of interest. Model comparison showed that subjective treatment did not lead to a better model fit neither as a main effect (df = 13, BIC = 1084.64, AIC = 1045.77, p = 0.069) nor as a two-way interaction with time (d f = 15, BIC = 1094.60, AIC = 1049.74, p = 0.990) and a three-way interaction with time and objective treatment (df = 21, BIC = 1122.48, AIC = 1059.68, p = 0.914). On the contrary, we next examined whether the changes in WM accuracy were explained by objective treatment. Our results showed that the inclusion of objective treatment (df = 11, BIC = 1089.73, AIC = 1056.84, p = 0.183) did not lead to a better model fit, but the interaction of objective treatment * time did (df = 15, BIC = 1094.60, AIC = 1049.74, p = 0.005). Therefore, when accounting for participants’ subjective experience of receiving the real or placebo treatment, the actual treatment to which participants were assigned during the experiment contributed to explaining changes in WM accuracy.

This study includes analyses on four datasets (Blumberger et al., 2016; Kaster et al., 2018; Leffa et al., 2022; Filmer et al., 2019). The dataset by Filmer et al., 2019 was originally open at: https://espace.library.uq.edu.au/view/UQ:3b30c6d. We made the de-identified datasets from Blumberger et al., 2016, Kaster et al., 2018 and Leffa et al., 2022 available on the Open Science Framework (OSF) upon approval from the study authors. The data and the R code can be found on the OSF at: https://osf.io/rztxu/.

1. 1. Hochman S
   2. Fassi L
   3. Cohen Kadosh R

   (2024) Open Science Framework

   The Importance of Accounting for Participants’ Subjective Beliefs When Assessing Outcomes in Neurostimulation Treatments.

   https://doi.org/10.17605/OSF.IO/RZTXU

1. 1. Filmer H
   2. Griffin A
   3. Dux PE

   (2019) UQ eSpace

   Dosage dependent increases in mind wandering via prefrontal tDCS.

   https://doi.org/10.14264/uql.2019.295

1. 1. Bates D
   2. Mächler M
   3. Bolker B
   4. Walker S

   (2015) Fitting linear mixed-effects models using lme4

   Journal of Statistical Software 67:i01.

   https://doi.org/10.18637/jss.v067.i01

   • Google Scholar
2. 1. Blumberger DM
   2. Maller JJ
   3. Thomson L
   4. Mulsant BH
   5. Rajji TK
   6. Maher M
   7. Brown PE
   8. Downar J
   9. Vila-Rodriguez F
   10. Fitzgerald PB
   11. Daskalakis ZJ

   (2016) Unilateral and bilateral MRI-targeted repetitive transcranial magnetic stimulation for treatment-resistant depression: a randomized controlled study

   Journal of Psychiatry & Neuroscience 41:E58–E66.

   https://doi.org/10.1503/jpn.150265

   • PubMed
   • Google Scholar
3. 1. Boot WR
   2. Simons DJ
   3. Stothart C
   4. Stutts C

   (2013) The pervasive problem with placebos in psychology: Why active control groups are not sufficient to rule out placebo effects

   Perspectives on Psychological Science 8:445–454.

   https://doi.org/10.1177/1745691613491271

   • PubMed
   • Google Scholar
4. 1. Braga M
   2. Barbiani D
   3. Emadi Andani M
   4. Villa-Sánchez B
   5. Tinazzi M
   6. Fiorio M

   (2021) The role of expectation and beliefs on the effects of non-invasive brain stimulation

   Brain Sciences 11:1526.

   https://doi.org/10.3390/brainsci11111526

   • PubMed
   • Google Scholar
5. 1. Cohen Kadosh R
   2. Cohen Kadosh K
   3. Schuhmann T
   4. Kaas A
   5. Goebel R
   6. Henik A
   7. Sack AT

   (2007) Virtual dyscalculia induced by parietal-lobe TMS impairs automatic magnitude processing

   Current Biology 17:689–693.

   https://doi.org/10.1016/j.cub.2007.02.056

   • PubMed
   • Google Scholar
6. 1. Connolly KR
   2. Helmer A
   3. Cristancho MA
   4. Cristancho P
   5. O’Reardon JP

   (2012) Effectiveness of transcranial magnetic stimulation in clinical practice post-FDA approval in the United States: results observed with the first 100 consecutive cases of depression at an academic medical center

   The Journal of Clinical Psychiatry 73:e567–e573.

   https://doi.org/10.4088/JCP.11m07413

   • PubMed
   • Google Scholar
7. 1. Davis NJ
   2. Gold E
   3. Pascual-Leone A
   4. Bracewell RM

   (2013) Challenges of proper placebo control for non-invasive brain stimulation in clinical and experimental applications

   The European Journal of Neuroscience 38:2973–2977.

   https://doi.org/10.1111/ejn.12307

   • PubMed
   • Google Scholar
8. 1. De Ridder D
   2. Maciaczyk J
   3. Vanneste S

   (2021) The future of neuromodulation: smart neuromodulation

   Expert Review of Medical Devices 18:307–317.

   https://doi.org/10.1080/17434440.2021.1909470

   • PubMed
   • Google Scholar
9. 1. Dubljević V
   2. Saigle V
   3. Racine E

   (2014) The rising tide of tDCS in the media and academic literature

   Neuron 82:731–736.

   https://doi.org/10.1016/j.neuron.2014.05.003

   • PubMed
   • Google Scholar
10. 1. Duecker F
    2. Sack AT

    (2015) Rethinking the role of sham TMS

    Frontiers in Psychology 6:210.

    https://doi.org/10.3389/fpsyg.2015.00210

    • PubMed
    • Google Scholar
11. 1. Filmer HL
    2. Griffin A
    3. Dux PE

    (2019) For a minute there, I lost myself … dosage dependent increases in mind wandering via prefrontal tDCS

    Neuropsychologia 129:379–384.

    https://doi.org/10.1016/j.neuropsychologia.2019.04.013

    • PubMed
    • Google Scholar
12. 1. Filmer HL
    2. Mattingley JB
    3. Dux PE

    (2020) Modulating brain activity and behaviour with tDCS: Rumours of its death have been greatly exaggerated

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 123:141–151.

    https://doi.org/10.1016/j.cortex.2019.10.006

    • PubMed
    • Google Scholar
13. 1. Fonteneau C
    2. Mondino M
    3. Arns M
    4. Baeken C
    5. Bikson M
    6. Brunoni AR
    7. Burke MJ
    8. Neuvonen T
    9. Padberg F
    10. Pascual-Leone A
    11. Poulet E
    12. Ruffini G
    13. Santarnecchi E
    14. Sauvaget A
    15. Schellhorn K
    16. Suaud-Chagny MF
    17. Palm U
    18. Brunelin J

    (2019) Sham tDCS: A hidden source of variability? Reflections for further blinded, controlled trials

    Brain Stimulation 12:668–673.

    https://doi.org/10.1016/j.brs.2018.12.977

    • PubMed
    • Google Scholar
14. 1. Fritsch B
    2. Reis J
    3. Martinowich K
    4. Schambra HM
    5. Ji Y
    6. Cohen LG
    7. Lu B

    (2010) Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning

    Neuron 66:198–204.

    https://doi.org/10.1016/j.neuron.2010.03.035

    • PubMed
    • Google Scholar
15. 1. Gandiga PC
    2. Hummel FC
    3. Cohen LG

    (2006) Transcranial DC stimulation (tDCS): A tool for double-blind sham-controlled clinical studies in brain stimulation

    Clinical Neurophysiology 117:845–850.

    https://doi.org/10.1016/j.clinph.2005.12.003

    • PubMed
    • Google Scholar
16. 1. Gordon MS
    2. Seeto JXW
    3. Dux PE
    4. Filmer HL

    (2022) Intervention is a better predictor of tDCS mind-wandering effects than subjective beliefs about experimental results

    Scientific Reports 12:13110.

    https://doi.org/10.1038/s41598-022-16545-0

    • PubMed
    • Google Scholar
17. 1. Grover S
    2. Nguyen JA
    3. Viswanathan V
    4. Reinhart RMG

    (2021) High-frequency neuromodulation improves obsessive-compulsive behavior

    Nature Medicine 27:232–238.

    https://doi.org/10.1038/s41591-020-01173-w

    • PubMed
    • Google Scholar
18. 1. Guerra A
    2. López-Alonso V
    3. Cheeran B
    4. Suppa A

    (2020) Variability in non-invasive brain stimulation studies: Reasons and results

    Neuroscience Letters 719:133330.

    https://doi.org/10.1016/j.neulet.2017.12.058

    • PubMed
    • Google Scholar
19. 1. Haikalis NK
    2. Hooyman A
    3. Wang P
    4. Daliri A
    5. Schaefer SY

    (2023) Placebo effects of transcranial direct current stimulation on motor skill acquisition

    Neuroscience Letters 814:137442.

    https://doi.org/10.1016/j.neulet.2023.137442

    • PubMed
    • Google Scholar
20. 1. Hallett M

    (2007) Transcranial magnetic stimulation: A primer

    Neuron 55:187–199.

    https://doi.org/10.1016/j.neuron.2007.06.026

    • PubMed
    • Google Scholar
21. 1. Hashmi JA

    (2018) Placebo effect: Theory, mechanisms and teleological roots

    International Review of Neurobiology 139:233–253.

    https://doi.org/10.1016/bs.irn.2018.07.017

    • PubMed
    • Google Scholar
22. 1. Horvath JC
    2. Forte JD
    3. Carter O

    (2015) Quantitative review finds no evidence of cognitive effects in healthy populations from single-session transcranial direct current stimulation (tDCS)

    Brain Stimulation 8:535–550.

    https://doi.org/10.1016/j.brs.2015.01.400

    • PubMed
    • Google Scholar
23. 1. Kaster TS
    2. Daskalakis ZJ
    3. Noda Y
    4. Knyahnytska Y
    5. Downar J
    6. Rajji TK
    7. Levkovitz Y
    8. Zangen A
    9. Butters MA
    10. Mulsant BH
    11. Blumberger DM

    (2018) Efficacy, tolerability, and cognitive effects of deep transcranial magnetic stimulation for late-life depression: A prospective randomized controlled trial

    Neuropsychopharmacology 43:2231–2238.

    https://doi.org/10.1038/s41386-018-0121-x

    • PubMed
    • Google Scholar
24. 1. Khedr EM
    2. Ahmed MA
    3. Fathy N
    4. Rothwell JC

    (2005) Therapeutic trial of repetitive transcranial magnetic stimulation after acute ischemic stroke

    Neurology 65:466–468.

    https://doi.org/10.1212/01.wnl.0000173067.84247.36

    • PubMed
    • Google Scholar
25. 1. Laird NM
    2. Ware JH

    (1982)

    Random-effects models for longitudinal data

    Biometrics 38:963–974.

    • PubMed
    • Google Scholar
26. 1. Lampit A
    2. Hallock H
    3. Valenzuela M

    (2014) Computerized cognitive training in cognitively healthy older adults: a systematic review and meta-analysis of effect modifiers

    PLOS Medicine 11:e1001756.

    https://doi.org/10.1371/journal.pmed.1001756

    • PubMed
    • Google Scholar
27. 1. Leffa DT
    2. Grevet EH
    3. Bau CHD
    4. Schneider M
    5. Ferrazza CP
    6. da Silva RF
    7. Miranda MS
    8. Picon F
    9. Teche SP
    10. Sanches P
    11. Pereira D
    12. Rubia K
    13. Brunoni AR
    14. Camprodon JA
    15. Caumo W
    16. Rohde LA

    (2022) Transcranial direct current stimulation vs sham for the treatment of inattention in adults with attention-deficit/hyperactivity disorder: The tuned randomized clinical trial

    JAMA Psychiatry 79:847–856.

    https://doi.org/10.1001/jamapsychiatry.2022.2055

    • PubMed
    • Google Scholar
28. 1. Looi CY
    2. Cohen Kadosh R

    (2016) Brain stimulation, mathematical, and numerical training: Contribution of core and noncore skills

    Progress in Brain Research 227:353–388.

    https://doi.org/10.1016/bs.pbr.2016.04.009

    • PubMed
    • Google Scholar
29. 1. López-Alonso V
    2. Cheeran B
    3. Río-Rodríguez D
    4. Fernández-Del-Olmo M

    (2014) Inter-individual variability in response to non-invasive brain stimulation paradigms

    Brain Stimulation 7:372–380.

    https://doi.org/10.1016/j.brs.2014.02.004

    • PubMed
    • Google Scholar
30. 1. McGough JJ
    2. Sturm A
    3. Cowen J
    4. Tung K
    5. Salgari GC
    6. Leuchter AF
    7. Cook IA
    8. Sugar CA
    9. Loo SK

    (2019) Double-blind, sham-controlled, pilot study of trigeminal nerve stimulation for attention-deficit/hyperactivity disorder

    Journal of the American Academy of Child and Adolescent Psychiatry 58:403–411.

    https://doi.org/10.1016/j.jaac.2018.11.013

    • PubMed
    • Google Scholar
31. 1. Medina J
    2. Cason S

    (2017) No evidential value in samples of transcranial direct current stimulation (tDCS) studies of cognition and working memory in healthy populations

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 94:131–141.

    https://doi.org/10.1016/j.cortex.2017.06.021

    • PubMed
    • Google Scholar
32. 1. Mikellides G
    2. Michael P
    3. Psalta L
    4. Stefani A
    5. Schuhmann T
    6. Sack AT

    (2022) Accelerated intermittent theta burst stimulation in smoking cessation: Placebo effects equal to active stimulation when using advanced placebo coil technology

    Frontiers in Psychiatry 13:892075.

    https://doi.org/10.3389/fpsyt.2022.892075

    • PubMed
    • Google Scholar
33. 1. Murphy OW
    2. Hoy KE
    3. Wong D
    4. Bailey NW
    5. Fitzgerald PB
    6. Segrave RA

    (2020) Transcranial random noise stimulation is more effective than transcranial direct current stimulation for enhancing working memory in healthy individuals: Behavioural and electrophysiological evidence

    Brain Stimulation 13:1370–1380.

    https://doi.org/10.1016/j.brs.2020.07.001

    • PubMed
    • Google Scholar
34. 1. Nitsche MA
    2. Paulus W

    (2000) Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation

    The Journal of Physiology 527 Pt 3:633–639.

    https://doi.org/10.1111/j.1469-7793.2000.t01-1-00633.x

    • PubMed
    • Google Scholar
35. 1. Oken BS
    2. Flegal K
    3. Zajdel D
    4. Kishiyama S
    5. Haas M
    6. Peters D

    (2008) Expectancy effect: impact of pill administration on cognitive performance in healthy seniors

    Journal of Clinical and Experimental Neuropsychology 30:7–17.

    https://doi.org/10.1080/13803390701775428

    • PubMed
    • Google Scholar
36. 1. Parkin BL
    2. Ekhtiari H
    3. Walsh VF

    (2015) Non-invasive human brain stimulation in cognitive neuroscience: A primer

    Neuron 87:932–945.

    https://doi.org/10.1016/j.neuron.2015.07.032

    • PubMed
    • Google Scholar
37. 1. Parong J
    2. Seitz AR
    3. Jaeggi SM
    4. Green CS

    (2022) Expectation effects in working memory training

    PNAS 119:e2209308119.

    https://doi.org/10.1073/pnas.2209308119

    • PubMed
    • Google Scholar
38. 1. Polanía R
    2. Nitsche MA
    3. Ruff CC

    (2018) Studying and modifying brain function with non-invasive brain stimulation

    Nature Neuroscience 21:174–187.

    https://doi.org/10.1038/s41593-017-0054-4

    • PubMed
    • Google Scholar
39. 1. Rabipour S
    2. Wu AD
    3. Davidson PSR
    4. Iacoboni M

    (2018) Expectations may influence the effects of transcranial direct current stimulation

    Neuropsychologia 119:524–534.

    https://doi.org/10.1016/j.neuropsychologia.2018.09.005

    • PubMed
    • Google Scholar
40. Software

    1. R Development Core Team

    (2022) R: A Language and Environment for Statistical Computing

    R Foundation for Statistical Computing, Vienna, Austria.

    https://www.R-project.org/
41. 1. Santarnecchi E
    2. Brem A-K
    3. Levenbaum E
    4. Thompson T
    5. Cohen Kadosh R
    6. Pascual-Leone A

    (2015) Enhancing cognition using transcranial electrical stimulation

    Current Opinion in Behavioral Sciences 4:171–178.

    https://doi.org/10.1016/j.cobeha.2015.06.003

    • Google Scholar
42. 1. Schmidt L
    2. Braun EK
    3. Wager TD
    4. Shohamy D

    (2014) Mind matters: placebo enhances reward learning in Parkinson’s disease

    Nature Neuroscience 17:1793–1797.

    https://doi.org/10.1038/nn.3842

    • PubMed
    • Google Scholar
43. 1. Sitaram R
    2. Ros T
    3. Stoeckel L
    4. Haller S
    5. Scharnowski F
    6. Lewis-Peacock J
    7. Weiskopf N
    8. Blefari ML
    9. Rana M
    10. Oblak E
    11. Birbaumer N
    12. Sulzer J

    (2017) Closed-loop brain training: the science of neurofeedback

    Nature Reviews. Neuroscience 18:86–100.

    https://doi.org/10.1038/nrn.2016.164

    • PubMed
    • Google Scholar
44. 1. Sonawalla SB
    2. Rosenbaum JF

    (2002) Placebo response in depression

    Dialogues in Clinical Neuroscience 4:105–113.

    https://doi.org/10.31887/DCNS.2002.4.1/ssonawalla

    • PubMed
    • Google Scholar
45. 1. Thibault RT
    2. Lifshitz M
    3. Raz A

    (2017) Neurofeedback or neuroplacebo?

    Brain 140:862–864.

    https://doi.org/10.1093/brain/awx033

    • PubMed
    • Google Scholar
46. 1. van Bueren NER
    2. Reed TL
    3. Nguyen V
    4. Sheffield JG
    5. van der Ven SHG
    6. Osborne MA
    7. Kroesbergen EH
    8. Cohen Kadosh R

    (2021) Personalized brain stimulation for effective neurointervention across participants

    PLOS Computational Biology 17:e1008886.

    https://doi.org/10.1371/journal.pcbi.1008886

    • PubMed
    • Google Scholar
47. 1. Walsh V
    2. Cowey A

    (2000) Transcranial magnetic stimulation and cognitive neuroscience

    Nature Reviews. Neuroscience 1:73–79.

    https://doi.org/10.1038/35036239

    • PubMed
    • Google Scholar
48. 1. Wang J
    2. Wen JB
    3. Li XL

    (2018) No effect of transcranial direct current stimulation of the dorsolateral prefrontal cortex on short-term memory

    CNS Neuroscience & Therapeutics 24:58–63.

    https://doi.org/10.1111/cns.12779

    • PubMed
    • Google Scholar
49. 1. Westwood SJ
    2. Olson A
    3. Miall RC
    4. Nappo R
    5. Romani C

    (2017) Limits to tDCS effects in language: Failures to modulate word production in healthy participants with frontal or temporal tDCS

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 86:64–82.

    https://doi.org/10.1016/j.cortex.2016.10.016

    • PubMed
    • Google Scholar

Author details

1. Luisa Fassi

   1. MRC Cognition and Brain Sciences Unit, University of Cambridge, Cambridge, United Kingdom
   2. Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom
   3. Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing

   Contributed equally with

   Shachar Hochman

   For correspondence

   Luisa.Fassi@mrc-cbu.cam.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0520-6425

2. Shachar Hochman

   School of Psychology, University of Surrey, Surrey, United Kingdom

   Contribution

   Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing

   Contributed equally with

   Luisa Fassi

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8322-3255

3. Zafiris J Daskalakis

   Department of Psychiatry, University of California, San Diego, San Diego, United States

   Contribution

   Resources, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

4. Daniel M Blumberger

   Temerty Centre for Therapeutic Brain Intervention at the Centre for Addiction and Mental Health and Department of Psychiatry, Temerty Faculty of Medicine, University of Toronto, Toronto, Canada

   Contribution

   Conceptualization, Resources, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8422-5818

5. Roi Cohen Kadosh

   1. Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom
   2. School of Psychology, University of Surrey, Surrey, United Kingdom

   Contribution

   Conceptualization, Supervision, Investigation, Methodology, Project administration, Writing – review and editing

   For correspondence

   r.cohenkadosh@surrey.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5564-5469

• 986

  views

• 81

  downloads

• 0

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.