Attention is a cognitive and behavioral process that selectively focuses on individual aspects of subjective or objective information, allowing through voluntary top-down and automatic bottom-up mechanisms to selectively process or inhibit contents from the multiplicity of sensory inputs over different domains [1
]. Attention facilitates or impairs other cognitive functions, such as memory, language, problem solving, and reflects complex interactions of multiple independent systems distributed within the brain [4
Psychiatric disorders can also lead to attention deficits. Dysfunctions in attentional processes and selective set-shifting have been reported in depressed individuals [6
]. Schizophrenia presents with positive clinical features but also with negative clinical features, such as attentional deficits [7
]. In adult patients with attention deficit hyperactivity disorder (ADHD), cognitive disturbances are more pronounced than in the pediatric population [8
] and are most evident as deficiencies of executive functions and attention [9
]. In autism, the selective attention has been shown to be impaired even in situations where behavior is normal; especially a deficit in rapid attention shifting has been observed in behavioral tasks shifting between sensory modalities, spatial locations, and object features [11
Attention does not localize anatomically [14
] and is therefore difficult to study. However, frontal regions are particularly active during tasks of alerting attention [5
]. Indeed, neuroimaging studies have demonstrated the engagement of the left dorsolateral prefrontal cortex (DLPFC) in executive functioning, and more specifically during selective attention. In particular, a functional magnetic resonance imaging (fMRI) study indicates the posterior DLPFC was active during a bimodal divided attention condition [15
]. The posterior DLPFC may support the increased working memory load associated with divided, compared to selective attention.
If delivered repetitively, transcranial magnetic stimulation (TMS) can influence brain function and induce changes in neuroplasticity, also in brain regions recruited by attentional processes. Indeed, repetitive TMS (rTMS) can modulate cortical excitability, inducing lasting effects [16
]. Therefore, rTMS has evolved into a powerful neuroscientific tool allowing to interfere transiently with specific brain functions.
A number of rTMS studies which targeted the DLPFC have shown significant improvements in cognitive function scores using both short- and long-term stimulation paradigms [17
]. It might be of interest to explore whether rTMS could serve as an intervention in disorders with attention deficits. A number of studies has specifically targeted attention, while many others assessed broader effects.
The aim of this review was to summarize the most specific studies assessing the effects of rTMS over DLPFC on attentional processes in subjects with psychiatric disorders.
2. Transcranial Magnetic Stimulation
rTMS is a noninvasive and safe brain stimulation technique that uses brief, intense pulses of electric current delivered to a coil placed on the subject’s head in order to generate an electric field in the brain via electromagnetic induction. rTMS has been proven to influence cortical excitability and the metabolic activity of neurons. Indeed, the induced electrical field modulates the neural transmembrane potentials and, thereby, neural activity. These effects depend on the intensity, frequency, and number of pulses applied, the duration of the course, the coil location and the type of coil used. RTMS can be applied as continuous trains of low-frequency (LF, 1 Hz) or bursts of higher frequency (HF, ≥5 Hz) rTMS. In general, LF rTMS is thought to reduce, and HF rTMS is thought to enhance excitability in the targeted cortical region [22
]. The physiological impact of rTMS and other neuromodulatory techniques involves synaptic plasticity, specifically long-term potentiation and long-term depression.
However, standard coils used in research and the clinic for rTMS are not capable of directly stimulating deep brain regions. The Heased coil (H-coil) is likely to have the ability of deep brain stimulation without the need of increasing the intensity to extreme levels [25
]. Deep TMS (dTMS) thus enables deeper noninvasive cortical stimulation at an effective depth of approximately 3 cm depending on the coil’s design and the stimulation intensity.
There is a sufficient body of evidence to accept with level of recommendation A (definite efficacy, Evidence Based Health Care) the analgesic effect of HF rTMS applied over the primary motor cortex contralateral to pain and the antidepressant effect of HF rTMS applied over the DLPFC [24
]. Overall, rTMS techniques have been shown to have potential therapeutic efficacy in cognitive neuroscience [26
]. In turn, these techniques have attracted worldwide attention as possible therapeutic tools for various neurological and psychiatric conditions [24
3. Material and Methods
In order to identify relevant articles for this review, we searched the MEDLINE, accessed by PubMed (1966–August 2018) and EMBASE (1980–August 2018) electronic databases were searched using the medical subject headings (MeSH) and free terms: “repetitive transcranial magnetic stimulation” OR “rTMS” AND “attention” OR “attentional” OR “attentive” AND “dorsolateral prefrontal cortex” OR “DLPFC”. Only original research articles were considered eligible for inclusion. Review articles or single case reports were excluded. The search was limited to studies written in English. Studies that met the following criteria were included: rTMS was conducted to patients with psychiatric diseases or neurological disorders with behavioral symptoms; administration site of rTMS was the DLPFC; the effect of rTMS on the cognitive domain attention was examined. In contrast, rTMS studies with animals as well as studies in which rTMS stimulation was administered on sites other than the DLPFC were excluded. Moreover, we included only studies that focused exclusively on attention, while studies with a broader scope within the umbrella concept of executive functions were excluded.
Full-text articles were retrieved for the selected titles, and reference lists of the retrieved articles were searched for additional publications. When data was missing or incomplete, principal investigators of included trials were contacted and additional information was requested. The titles and abstracts of the initially identified studies were screened by two authors to determine whether they satisfied the selection criteria. The methodological quality of each study and risk of bias were independently assessed, focusing on blinding, and any disagreement was solved through discussion. This search strategy yielded 24 results, three of which were excluded after reading the full paper, thereby leaving 21 studies which contributed to this review.
A flow-chart (Figure 1
) shows the selection/inclusion process.
This review highlights that rTMS applied over the DLPFC can positively influence the attentional function in subjects with several psychiatric disorders. The outcome measures were not uniform but mostly dealt with attentional performance.
Some studies revealed that prefrontal rTMS could exert procognitive effects on executive function and attention in patients with depression [3
]. Antidepressant effects of rTMS could be related to the same neurochemical mechanisms that underlie cognitive functioning, or just facilitate the normal cognitive function that was repressed because of the severe effects depression has on overall physical and cognitive functioning. It has been hypothesized that the extent of antidepressant effects could be considered as second-order long-term effects possibly related to primary alternations in cognitive functioning. Concurrence of depression and cognitive dysfunctions is well known in a wide range of clinical populations [61
]. In particular, impaired cognition is closely related to depressive symptoms in AD [62
], thus possibly potentiating the devastating effects of the disease itself or being an early sign of neural dysfunction [64
In patients with schizophrenia, imaging studies have demonstrated abnormalities in the left globus pallidus, which lead to widespread hypometabolism affecting the frontal lobes, especially the DLPFC and the anterior cingulate gyrus [65
]. Furthermore, abnormalities of visually orienting the frontal lobes/executive attentional network could interact with the parietal lobes/orienting network to affect the initiation of attentional shift, thus leading to abnormalities of visual orienting [66
]. It is therefore of interest that rTMS to the DLPFC could improve attentional functioning in this patient population [38
]. However, the findings were contradictory, as other studies could not identify any beneficial effects. A more systematic investigation comparing the different parameters of TMS to each other may shed more light on the mechanisms of action.
The results of some studies support the use of LF rTMS as a modulatory tool to alter the disrupted balance between cortical excitation and inhibition in autism. LF rTMS application to DLPFC would result in an alteration of the abnormal excitatory/inhibitory ratio through the activation of inhibitory GABAergic double bouquet interneurons.
Similarly, in patients suffering from ADHD initial findings suggest the possibility that attentional difficulties can be improved by using HF rTMS applied to the right DLPFC, and have encouraged future research [41
]. However, the evidence from a more recent study does not support the effectiveness of bilateral prefrontal stimulation to treat adult ADHD [42
]. Due to the small sample size, these preliminary results should be interpreted with caution.
rTMS can significantly improve, among other cognitive functions, attentional impairment that often accompanies AD. Impairments in visual attention and visual information processing have been identified as part of the neuropsychological features of AD, even in its earliest stages, and dissociations in visual attention deficits have been detected also in mild cognitive impairment (MCI) using a measure that assesses simple, divided, and selective attention [67
]. It is unclear whether the memory impairment in patients with amnestic MCI (aMCI) and AD is associated with attentional deficits. An fMRI study revealed that there are changes in the functional network subserving divided attention in patients with aMCI, as reflected in the attenuated activation of PFC [68
]. Interestingly, depressive symptoms in AD patients increase the deficits of cognitive flexibility and divided attention [69
This review has some limitations. First of all, there is considerable variability between studies in patients with different neuropsychiatric diseases. Very few trials have used exactly the same study design. The stimulation protocols, with respect to frequency, intensity, orientation of the coil, pattern, number of pulses by train, total number of pulses, duration of stimulation, frequency and intensity of stimulation, number of sessions delivered, are highly heterogeneous. Therefore, estimating the real effectiveness and reproducibility is very difficult. Systematic investigation of the effects of the various stimulation protocols are highly warranted, because the border between effectiveness and ineffectiveness may be very small and occurs somewhere in the dimensions spanned by the abovementioned parameters.
Furthermore, we have included in this review only studies employing specific cognitive tests/tasks focusing on attention, even if working memory and other executive functions are strongly correlated with this cognitive domain. Indeed, the role of the right DLPFC and of the right posterior parietal cortex (PPC) in controlling the interaction between working memory and attention during a visual search has been explored using rTMS in a recent study [70
]. Both the rDLPFC and the right PPC were found to be critical for controlling working memory biases in human visual attention. However, the broader scope of including executive functions should be addressed in another systematic summarizing work; possibly a meta-analysis could be conducted given that the study protocols were more comparable.
It should be considered that most therapeutic attempts are based on rTMS techniques aiming at enhancing cortical excitability, in particular HF rTMS. However, the underlying pathophysiologic mechanisms differ among the various neurological and psychiatric diseases which can be treated with this noninvasive brain stimulation technique. Therefore, appropriate testing of cortical physiology before and after therapeutic interventions is needed.
In conclusion, a better understanding of attention networks could allow targeting the most suitable area of the brain according to the specific attention domain affected. Moreover, a detailed examination of the best stimulation frequency, surface or deep stimulation, duration and intensity of the intervention, among other important core features of TMS-protocols, should be done when moving closer to clinical application of TMS to treat attentional deficits.
Despite the above-mentioned limitations, this review indicates that neuromodulatory techniques such as rTMS are promising approaches to be used as attentional enhancers in people with neuropsychiatric conditions where impaired attention is a prominent feature.