Skip to main content
. 2020 May 6;11:378. doi: 10.3389/fpsyt.2020.00378

Table 2.

Summary of fNIRS studies differentiating depressed patients from healthy controls.

Source Country Sample size (male/female) Age (mean ± standard deviation) Diagnostic criteria (instrument) Psychopathology measure Medication NIRS device Paradigm Brain area Main findings
Pu et al., (25) Japan MDD: 24 (12/12)
HC: 26 (8/18)
MDD: 47.9 ± 13.9
HC: 42.4 ± 9.3
DSM-IV-TR (MINI) BDI
HAMD
All on antidepressants 52-Channel NIRS
(ETG-4000)
2-Back task with blocked periodic baseline, activation PF
T
– MDD group had ↓ response sensitivity and accuracy than HC.
Matsuo et al. (26) Japan UNI: 8 (1/7)
BP: 1 (1/0)
HC: 10 (0/10)
UNI+BP:
65.6 ± 6.4
HC: 59.5 ± 5.9
DSM-IV
(unspecified)
HAMD All on medication HEO-200 Verbal repetition task, VFT, Hyperventilation, Paper-bag breathing F – VFT: Oxy-Hb ↑ and deoxy-Hb ↓ in HC group but no noteworthy changes in depressed group.
– Hyperventilation: oxy-Hb ↓ while deoxy-Hb ↑
Herrmann et al. (27) Germany MDD: 9 (5/4)
HC: 9 (5/4)
MDD: 37.3 ± 13.8
HC: 35.1 ± 5.5
ICD-10 BDI All on medication 2-Channel NIRO-300 monitor VFT PF – Oxy-Hb ↑ in HC.
– MDD had significantly ↓ activation.
Shoji et al. (28) Japan MDD: 26
HC: 32
NA ICD-10 HAMD NA 44-channel ETG-4000 (Hitachi) Word generation tasks PF – Oxy-Hb variations in MDD were appreciably lesser than HC in all word tasks.
Kinoshita et al. (29) Japan MDD: 17 44.2 ± 12.2 DSM-IV (SCID) HAMD Majority on antidepressants 22-Channel ETG-4000 (Hitachi) DEX/CRH test
VFT
F – Results did not fit well with the diagnostic criteria (DSM or ICD).
Kito et al. (30) Japan MDD: 30 (9/21)
HC: 33 (11/22)
MDD: 71.1 ± 6.8
HC: 69.6 ± 5.5
DSM-IV HAMD All on medication FOIRE-3000 (Shimadzu) VFT F, P – Cortical activation in the VFT in MDD ↓ compared to HC.
Matsuo et al. (31) Japan MDD: 14 (4/10)
HC: 21 (3/18)
MDD: 56.1 ± 17.3
HC: 50.3 ± 12.6
DSM-IV HAMD All on medication Single channel HEO-200 (Omron) VFT
Hyperventilation
Paper-bag breathing
F – VFT: ↑ in oxy-Hb was lower in MDD compared to HC.
– Hyperventilation: MDD demonstrated an appreciably smaller reduction in oxy-Hb than HC.
– Paper-bag breathing: oxy-Hb ↑ in both groups while deoxy-Hb ↓ in MDD.
Matsubara et al. (32) Japan MDD: 10
HC: 10
NA NA NA NA 52-Channel ETG-4000 (Hitachi) Emotional Stroop task T – HC showed ↑ in oxy-Hb in the fronto-temporal regions as opposed to MDD.
Koike et al. (33) Japan MDD: 405
HC: 369
NA NA NA NA fNIRS VFT PF – The intensity of signals was smaller in MDD.
– The duration of time taken to complete the task was later in MDD.
Azechi et al. (34) Japan MDD: 30
HC: 30
NA NA NA NA 2-Channel NIRS VFT, TOH, SBT Stroop Task F – Task performances of the VFT and TOH were ↓ in MDD than in HC.
Matsuo et al. (35) Japan MDD: 10 (5/5)
HC: 10 (6/4)
MDD: 62.2 ± 4.8
HC: 58.7 ± 5.8
DSM-IV (MINI) HAMD Only 4 patients had medication 24-Channel ETG-100 (Hitachi) VFT
WRT
CO2 inhalation
PF – During cognitive task, there was ↓ activation of PF cortex in MDD.
– Negative association between ↓ PF activation while performing the cognitive task and degree of hyperintensity in the periventricular region or left F cortex in MDD.
– CO2 inhalation causing vasomotor reactivity was ↓ in MDD than HC.
Ohta et al. (36) Japan MDD: 17 (5/12)
HC: 24 (12/12)
MDD: 42.8 ± 18.2
HC: 36.2 ± 16.5
DSM-IV
(MINI)
HAMD All on medication 52-Channel ETG-4000 (Hitachi) WFT F – Improvement of oxy-Hb values in the bilateral F cortices,
– MDD showed attenuated ↑ in oxy-Hb while doing WFT in the bilateral F regions.
– Hypofrontality in MDD is most notable in the left medial inferior F lobe.
– Enhancement of oxy-Hb was not as notable on the right-sided channels in MDD.
Suto et al. (18) Japan MDD: 10
(9/1)
HC: 16
(12/4)
MDD: 47.9 ± 12.8
HC: 42.9 ± 4.6
DSM-IV HAMD All on medication 24-Channel ETG-100 (Hitachi) WFT
Right-finger-tapping task
F, T – During first half of task period, MDD had ↓ oxy-Hb increase than HC. ↑were seen in the anterior lower T and lower F channels.
– Finger tapping tasks in MDD patients caused ↑ in oxy-Hb compared to HC.
Shimodera et al. (37) Japan UNI: 39 (19/20)
BP: 14 (7/7)
HC: 24 (13/11)
UNI: 56.9 ± 12.6
BP: 51.4 ± 14.0
HC: 40.9 ± 10.6
DSM-IV-TR HAMD All on medication 52-Channel OMM-3000/16 (Shimadzu) VFT PF – Both UNI and BP showed ↓ area under curves (AUCs) than HC.
– MDD showed significantly ↓ weighted center (WC) than BP or HC.
Ma et al. (38) China Menopausal MD: 30 (0/30)
MDD: 30 (0/30)
HC: 30 (0/30)
MD: 51.17 ± 6.06
MDD: 37.50 ± 10.60
HC: 34.83 ± 8.77
DSM-IV HAMD Only 20 patients took medication 45-Channel FOIRE-3000 (Shimadzu) VFT PF – MD and MDD both showed ↓ oxy-Hb activation in bilateral DLPFC than HC, but involving different channels.
– Atypical hemodynamics of the left DLPFC can discriminate HC from MDD using NIRS.
Takizawa et al. (39) Japan UNI: 153 (76/77)
BP: 134 (65/69)
HC: 590 (276/314)
UNI: 43.8 ± 12.7
BP: 44.0 ± 14.9
HC: 43.9 ± 15.7
DSM-IV
SCID
HAMD 10 drug-free patients with UNI 52-Channel ETG-4000 (Hitachi) VFT PF, T – NIRS can differentiate HC from patients, and differentiate UNI from BP.
Liske et al. (40) Germany MDD: 41
HC: 46
NA NA SIMS NA NA Motor task–pressing button P – In HC, a noteworthy contrast in the NIRS signal resulting from the left P brain regions but this was weaker in MDD.
Zhu et al. (41) China UNI: 35 (11/24)
BP: 39 (19/20)
HC: 36 (18/18)
UNI: 35.9 ± 13.2
BP: 37.0 ± 12.9
HC: 33.6 ± 10.3
DSM-IV-TR (MINI) HAMD All on medication 52-Channel ETG-4000 (Hitachi) 1-Back version of the n-back WMT PF, T – Compared to HCs, UNI and BP ↓ activation of oxy-Hb in the inferior PF region during WMT.
– Distinct prefrontal activation patterns in the Broca's area and left frontopolar region, underline BD and UD.
Matsubara et al. (42) Japan UNI: 16 (8/8)
BP: 16 (8/8)
HC: 20 (10/10)
UNI: 45.4 ± 12.2
BP: 44.1 ± 17.5
HC: 41.4 ± 8.5
DSM-IV-TR (MINI) HAMD All on medication 52-Channel ETG-4000 (Hitachi) Emotional Stroop task PF -While doing the threat task, depressed patients revealed ↑ oxy-Hb in the left middle frontal region.
– When performing happy task, depressed patients, did not display any meaningful changes in oxy-Hb.
Akashi et al. (43) Japan MDD: 52 (32/20)
Subdivided into with/without discrepancy
HC: 48 (21/27)
MDD: 41.8 ± 12.7
HC: 38.9 ± 9.5
DSM-IV
(MINI)
Structured interview guide for HAMD (SIGH-D) Most of the patients taking medications 52-Channel ETG-4000 (Hitachi) VFT F, T – In the FT regions, upsurge in mean oxy-Hb were ↓ in MDD than in HC.
Gao et al. (44) China MDD: 27 (7/20)
HC: 24 (11/13)
MDD: 40.78 ± 13.42
HC: 43.13 ± 11.28
DSM-IV HAMD All were medication free for at least 4 weeks CW-NIRS Facial emotion recognition PF – Hemodynamic variations between the bilateral PF cortex and left PF cortex may provide dependable predictors for diagnosing depression
Schecklmann et al. (45) Germany UNI: 16 (9/7)
BP: 14 (3/11)
HC: 15 (7/8)
UNI: 43.4 ± 9.8
BP: 40.8 ± 10.2
HC: 40.9 ± 8.0
ICD-10 BDI-II All on medication 52-Channel ETG-4000 (Hitachi) WMT PF, F Results discovered unspecific deficits that inhibited discrimination between bipolar and unipolar depression in domains of working memory.
Tomioka et al. 3,4 (46) Japan MDD: 25 (3/22)
HC: 62 (14/48)
MDD: 51.9 ± 16.6
HC: 51.7 ± 17.2
DSM-IV HAMD Medication-naive 52-Channel ETG-4000 (Hitachi) VFT PF, T – MDD showed ↓ oxy-Hb values when comparing to with HC in the bilateral F and T cortices at baseline.
– Hypofrontality response to VFT may represent a potential trait marker for depression
Ohtani et al. 3,4 (47) Japan UNI: 10 (4/6)
BP: 18 (9/9)
HC: 14 (7/7)
UNI: 39.2 ± 12.1
BP: 39.7 ± 9
HC: 33.6 ± 8.3
DSM-IV-TR HAMD All on medication 52-Channel ETG-4000 (Hitachi) VFT PF, T – UNI and BP showed ↓ activation than HCs in the bilateral ventrolateral PF cortex and the anterior part of the T cortex.
Masuda et al. 3,4 (48) Japan MDD: 47 Response group to SSRIs: 28 (15/13)
Nonresponse group: 19 (6/13)
HC: 63 (35/28)
Response group: 48.9 ± 2.9)
Nonresponse group: 43.2 ± 3.3
HC: 41.7 ± 1.4
DSM-IV-TR HAMD Medication-naive 47-Channel ETG-7100 (Hitachi) VFT PF, T – In FT region, hemodynamic responses were ↓ in patients with response and nonresponse than in HC prior to treatment.
– In the medial F region, hemodynamic responses were ↑ in patients with response to prior treatment.
Feng et al. 4 (49) China MDD: 15 (7/8)
HC: 15 (6/9)
MDD: 30.93 ± 13.47
HC: 30.87 ± 10.11
DSM-5
SCI
HAMD NA 45-Channel FOIRE-3000 (Shimadzu) VFT PF – After the music treatment, average active oxy-Hb values of some channels were ↑ in both HC and MDD.
– After music therapy, patients with MDD demonstrated substantial activation in the OFC, DLPFC, and VMPFC.
Hirano et al. 3,4 (50) Japan MDD: 30 (11/19)
HC: 108 (45/63)
MDD: 59.4 ± 14.2
HC: 58.9 ± 13
ICD-10 MADRS
QIDS-SR
All on medication 52-Channel ETG-4000 (Hitachi) VFT PF, T – MDD exhibited ↓ oxy-Hb values in the bilateral F cortex while doing VFT than HC.
Downey et al. 4 (51) UK MDD: 18
HC: 51
NA NA NA NA MiniNTS 4 detectors/24 sources (UCL) VFT, N-back working memory tasks PF – MDD had ↓ bilateral PF cortex oxy-Hb responses to VFT unlike HC, and this was additionally ↓ after 4 ECT sessions.
– On WM task, MDD showed PF cortex inhibition at baseline and a different time course of oxy– and deoxy-Hb following 4 ECT.
Rosenbaum et al. 3 (52) Germany MDD: 60
HC: 24
MDD: 40 ± 14.79
HC: 33 ± 11.45
DSM-IV
(SCI)
PHQ-9
MADRS
32% of patients treated with antidepressant medication 52-Channel ETG-4000
(Hitachi)
7-min resting phase, VAS, rumination response scale P – MDD as compared to HC, showed ↓ functional connectivity in parts of the DMN.
Kondo et al. 3 (53) Japan MDD: 25 (17/8)
HC: 25 (18/7)
MDD: 36 ± 8.91
HC: 34.1 ± 10.1
DSM-IV-TR
SCI
HAMD All medicated with antidepressants 44-Channel ETG-4000 (Hitachi) Image-recall task F, T – The oxy-Hb in HC was ↑ compared to MDD in bilateral F region.
– The severity of depression was related to ↓ in oxy-Hb in left F lobe.
Zhu et al. 3 (54) China Affective disorder (AD): 28 (8/20)
HC: 30 (21/9)
AD: 23.32 ± 5.01
HC: 23.60 ± 2.03
DSM-IV SDS 13 patients were medicated.
15 free of any medicine
42-Channel
FOIRE-3000
(Shimadzu)
Resting state measurements F – Relative to HC, AD demonstrated ↓ in intraregional and symmetrical interhemispheric connectivity in the PFC, revealed ↓ locally functional connectivity in the right IFG, and ↓ long-distance connectivity concerning the bilateral IFG.
Okada et al. 3 (55) Japan MDD: 36 (24/12)
HC: 36 (21/12)
MDD (male): 23.3 ± 2.5
MDD (female): 21.3 ± 1.1
HC (male): 23.9 ± 2.4
HC (female): 23.6 ± 2.1
DSM-III-R HAMD 13 patients received antidepressant medication.
23 were medication free for a minimum of 3 months
Multichannel near-IR spectrophotometry (NIRS) MDT LT, RT brain – Nearly half of patients revealed a “nondominant hemisphere response pattern”
– The supposedly “nondominant” hemisphere may convert to being dominant during the depression
Uemura et al. 3 (56) Japan MDD: 13 (6/7)
HC: 67 (28/39)
MDD: 74.5 ± 5.8
HC: 73.8 ± 5.3
NA GDS All subjects were medicated. 8-Channel FOIRE– 3000; (Shimadzu) Trail-making test, part B (TMT-B; tablet version) PF – Oxy-Hb activation when performing the TMT-B was ↓ in MDD in both the right and left PF cortex.
– ↓ PF activation in the elderly with depressive symptomology may cause deterioration in executive function.
Kinou et al.3 (57) Japan MDD: 32 (15/17)
HC: 32 (15/17)
MDD: 44.8 ± 9.8
HC: 45.7 ± 13.5
DSM-IV H All, except 3 patients 52-Channel ETG-4000 (Hitachi) VFT PF – MDD revealed ↓ oxy-Hb activation during the task.
Yamagata et al.3 (58) Japan Early-onset depression (EOD): 11 (2/9)
Late-onset (LOD): 12 (3/9)
HC: 13 (8/5)
EOD: 68.4 ± 5.6
LOD: 70.2 ± 1.9
HC: 70.3 ± 4.4
DSM-IV HAMD All patients were taking one prescribed
antidepressant.
52-Channel ETG-4000
(Hitachi)
WFT F, T – HC demonstrated ↑ in oxy-Hb vs. surges in oxy-Hb being minimally ↓ in EOD and extremely ↓ in LOD, bilaterally throughout the F cortices and T areas.
– Attenuated activation in the left lateral PF and T areas may help to differentiate LOD and EOD.
Noda et al.3 (59) Japan MDD: 30 (14/16)
HC: 30 (14/16)
MDD: 36.7 ± 11.6
HC: 35.1 ± 9.4
DSM-IV
(SCI)
GRID-HAMD All patients
medicated with antidepressants.
52-Channels ETG-4000
(Hitachi)
VFT F. T – Oxy-Hb increases while performing task was meaningfully ↓ in MDD.
– ↓ Right F-T activation on NIRS during VFT is related to the MDD severity.
Nishizawa et al. (60) Japan MDD: 14 (7/7)
HC: 20 (13/7)
MDD: 38.2 ± 12.9
HC: 29.0 ± 5.7
DSM-IV-TR (SCID) HAMD All taking antidepressants except for four patients. 22-Channel ETG-4000 (Hitachi) Emotional Stroop task F, T – Hyperactivated oxy-Hb was witnessed in the left F cortex on contact to unfavorable stimuli, but no noteworthy dissimilarity was established between MDD and HC on exposure to favorable stimuli.
Akiyama
et al. 3 (61)
Japan MDD: 177 (73/104)
HC: 50 (40/10)
MDD: 47.2 ± 15.1
HC: 32.7 ± 7.5
DSM-IV-TR HAMD All patients were on antidepressants 52-Channel ETG-4000 (Hitachi) VFT PF, T – Significant hypoactivation in bilateral F-T regions was observed in MDD compared with HC.
Ohi et al.3 (62) Japan UNI: 26 (17/9)
BP: 22 (13/9)
HC: 51 (33/18)
UNI: 41.1 ± 12.7
BP: 39.9 ± 12.5
HC: 35.7 ± 11.9
DSM-V HAMD Chlorpromazine equivalents of total antipsychotics 52-Channel
ETG-4000 (Hitachi)
VFT PF – UNI and BP groups had ↓ PF activity than HC.
– Patients with and without family history had ↓ PF activity than HC subjects.
– Demonstrate connection of more serious PF dysfunction with higher genetic loading for disease.
Takei et al.3 (63) Japan UNI: 29 (14/15)
BP: 31 (14/17)
HC: 31 (11/20)
UNI: 34.5 ± 9.0
BP: 34.9 ± 6.6
HC: 33.6 ± 10.0
DSM-IV HAMD Nearly all patients
were on medication.
52-Channel
ETG-4000 (Hitachi)
Conversation task and control task F, T – Both UNI and BP showed ↓ of continuous activation in the left DLPFC and left FPC, and decreased rapid change in bilateral FPC activation.
– F activation while conversing ↓ in equally in UNI and BP.
– Pathophysiological character of UNI and BP are reflected in continuous activation and rapid change
change Impaired adaptive ability in UNI may be related to ↓ amount of rapid change in right FPC.
Pu et al.3 (64) Japan Late-onset depression (LOD): 24 (6/18)
HC: 30 (14/16)
LOD: 72.3 ± 5.5
HC: 72.0 ± 4.7
DSM-IV
(MINI)
BDI
HAMD
Antidepressant-naive 52-Channel
ETG-4000 (Hitachi)
VFT PF, T – LOD had ↓ activation in both PF and superior T cortices than HC.
– ↓ frontopolar cortical activation was linked with social functioning impairment in LOD.
Tsujii et al.3 (65) Japan MDD:
Suicide attempters (SAs): 30 (8/22)
Nonattempters (NAs): 38 (16/22)
HC: 40 (15/25)
MDD (SAs): 37.6 ± 10.0
MDD (NAs): 38.8 ± 9.7
HC: 38.2 ± 10.5
DSM-IV
(MINI)
HAMD All on medication. 52-Channel ETG-4000 (Hitachi) VFT F, T – MDD had significantly ↓ activation in the bilateral FT regions compared to HCs.
– SAs demonstrated ↓ hemodynamic response in the left precentral gyrus than HCs and NAs.
– Aggression and hopelessness were negatively associated with hemodynamic responses in the right middle F gyrus in SAs but not in HCs and NAs.
Pu et al.3 (66) Japan Late-onset depression (LOD): 36 (9/27)
HC: 35 (11/24)
LOD: 71.8 ± 5.1
HC: 70.9 ± 4.3
DSM-IV
(MINI)
BDI
HAMD
Antidepressant-naive 52-Channel ETG-4000
(Hitachi)
Working memory (WM) task PF, T – LOD was correlated with ↓ PF and T activation when comparing with HC.
– Hemodynamic response in PF and T regions when performing WM task could correlate with social functioning in LOD.
Tsujii et al.3 (67) Japan MDD with melancholia (MDD-MF): 30 (15/15)
MDD without -melancholia (MDD-NMF): 52 (18/34)
HC: 68 (32/36)
MDD-MF: 42.2 ± 11.8
MDD-NMF: 40.6 ± 11.7
HC: 40.5 ± 10.6
DSM-IV (MINI) HAMD All on medication 52-Channel ETG-4000
(Hitachi)
VFT F, T – Both MDD groups demonstrated ↓ hemodynamic responses in the F-T regions.
Liu et al.3 (68) China MDD: 30 (12/18)
HC: 30 (16/14)
MDD: 38.38 ± 12.8
HC: 33.2 ± 10.5
DSM-IV-TR HAMD Free of medication 52-Channel FOIRE-3000
(Shimadzu)
VFT PF – ↓ Activation in lateral and lower PFC in MDD.
– Antero-medial PFC and bilateral PFC were correlated with the degree of depressive symptoms.
– MDD patients with obsession–compulsion symptoms and anxiety exhibited a PFC ↓ activation state in NIRS.
Wang et al.3 (69) China First-episode MDD (fMDD): 36 (15/21)
Recurrent MDD: 34 (11/23)
HC: 37 (22/15)
fMDD: 38.75 ± 13.86
Recurrent MDD: 43.26 ± 13.85
HC: 35.70 ± 11.39
DSM-IV HAMD All were on antidepressants 52-Channel ETG-4000 (Hitachi) VFT F, T – In comparison with HC and fMDD, chronic MDD had significantly ↓ brain activation over right PF and superior T cortices.
– Variation in activations in bilateral F and T regions. HC: more channels in left than right hemisphere; fMDD: more channels in right than left hemisphere; recurrent MDD: only channels in left hemisphere.
Tsujii et al.3 (70) Japan MDD with melancholia (MDD-MF):
32 (16/16)
MDD without melancholia (MDD-NMF):
28 (15/13)
HC: 24 (11/13)
MDD-MF:
40.8 ± 15.3
MDD-NMF:
38.9 ± 11.8
HC: 38.6 ± 9.2
DSM-IV (MINI) SIGH-D
BDI-II
All on medication 52-Channel ETG-4000
(Hitachi)
VFT F, T – Noteworthy disparities were witnessed in mean oxy-Hb fluctuations of MDD-MF in 25 channels and in those with MDD-NMF in 12 channels compared to HC.
Nishida et al.3 (71) Japan MDD: 14 (7/7)
HC: 15 (8/7)
MDD: 46.2 ± 11.9
HC: 45.5 ± 10.9
DSM-IV-TR (MINI) HAMD All on medication 52-Channel ETG-4000 (Hitachi) VFT PF, T – MDD revealed a ↓ oxy-Hb activation than HC, predominantly in ventrolateral PF and T cortex regions.
Rosenbaum et al.3 (72) Germany Depressed: 49
Nondepressed: 51
2 participants diagnosed with bipolar disorder and eating disorder.
Depressed:
64.08 ± 7.06
Nondepressed:
64.16 ± 6.14
NA BDI
GDS
54% took medication. 38-Channel
ETG-4000 (Hitachi)
Adapted trail-making test (TMT-A, TMT-B, and TMT-C) F, P – Depressed and nondepressed revealed substantial differences in functional connectivity (FC) while doing the task performance and at rest.
– During task performance, depressed patients exhibited ↓ FC in a left frontopolar cortical network, and ↑ FC in a left frontoparietal cortical network at the resting state and altered FC and network organization during different mental states.
Pu et al.3
(73)
Japan MDD: 67 (29/38)
MDD with suicidal ideation: 31 (11/20)
MDD without suicidal ideation: 36 (18/18)
HC: 67 (29/38)
MDD: 58.1 ± 16.0
MDD with suicidal ideation: 57.3 ± 15.7
MDD without suicidal ideation: 58.7 ± 16.5
HC: 58.1 ± 17.8
DSM-IV
(MINI)
HAMD All on antidepressants 52-Channel ETG-4000 (Hitachi) VFT PF, T – Regional hemodynamic changes were considerably ↓ in MDD than in HCs in PF and T regions.
Pu et al.3
(74)
Japan MDD: 26 (11/15)
HC: 30 (12/18)
MDD: 47.9 ± 19.2
HC: 50.5 ± 19.7
DSM-IV-TR
(MINI)
BDI
HAMD
All on antidepressants 52-Channel ETG-4000
(Hitachi)
VFT PF, T – Regional hemodynamic changes were appreciably ↓ in MDD than in HC in PF and T regions, and was correlated positively with task-oriented coping (adaptive coping) in the bilateral ventrolateral and dorsolateral prefrontal cortex, and the midline frontopolar and bilateral orbitofrontal cortex regions.

3This article can also be found in the summary in Table 3.

4This article can also be found in the summary in Table 4.

MDD, major depressive disorder; HC, healthy control; BDI, Beck depression inventory; HAMD, Hamilton depression rating scale; UNI, unipolar depression; BP, bipolar disorder; VFT, verbal fluency task; PF, prefrontal; T, temporal, F, frontal; NA, not available; TOH, Tower of Hanoi; SBT, Sternberg's task; WRT, word repetition task; MINI, Mini-international neuropsychiatric interview; DSM, Diagnostic and Statistical Manual; ICD, International Classification of Disease; SCID, structured clinical interview for DSM-IV; SIMS, structured inventory of malingered symptomatology; WMT, working memory task; IFG, inferior frontal gyrus; MDT, mirror drawing task; DLPFC, dorsolateral prefrontal cortex; PFC, prefrontal cortex; SIGH-D, structured interview guide for the Hamilton depression rating scale.