Dead bug

Movement

What the movement is

The dead bug is done lying face-up, lowering an opposite arm and leg toward the floor. The studies on this page do not share one form of it^2,9,10:

• A table-top position. In an ultrasound study of Pilates exercises, participants lay on the back with the arms straight over the chest². The hips and knees were bent to 90 degrees, with the feet off the ground and the lower back touching the mat². They did the exercise for ten seconds with the abdominal drawing-in manoeuvre, drawing the belly button in toward the spine².
• An arm and the opposite leg held low. A published trial protocol for 32 male athletes with chronic (long-lasting) low back pain also lists the dead bug with the drawing-in manoeuvre⁹. There one arm and the leg on the other side are held just above the horizontal⁹.
• Moving limbs. A randomised study of 45 participants with chronic non-specific low back pain used a “dead bug track”¹⁰. It began supine (lying face-up) with the knees bent and the arms at the sides¹⁰. It moved on to coordinated opposite and alternating arm and leg movements with diagonal holds¹⁰.

A related test: supine leg lowering

Some findings on this page come from a different movement, the Sahrmann five-level core stability test, and not from the dead bug¹¹. It is a set of supine leg-lowering manoeuvres of rising difficulty¹¹. They start with the hips and knees bent to about 90 degrees¹¹. One leg or both legs are then lowered and slid out straight¹¹. The heel either touches the surface or is held about 12 cm above it¹¹.

What it loads

Which abdominal muscles were most active

One study used electromyography (EMG), which measures a muscle’s activity by picking up its electrical signals¹. Its 19 healthy participants and 13 with low back pain did Pilates mat exercises¹. The dead bug activated the obliques more than the multifidus and rectus abdominis¹.

Of the three mat exercises in that study, one named the criss-cross gave the highest rectus abdominis and oblique activity¹. The dead bug and an exercise named the single leg stretch came next, and were similar to each other¹.

A study of an exercise named the Dying Bug

One study of 12 healthy subjects tested an exercise its authors call the Dying Bug¹². This page reports it apart from the dead bug studies. The exercise mainly recruited the abdominal muscles¹². The rectus abdominis and the abdominal oblique were equally active, and their activity rose in proportion as the exercise was made more difficult¹².

No muscle measured went above 41% of maximum voluntary isometric contraction, a muscle’s hardest effort while the joint stays still¹². The authors suggest that this intensity is not likely enough for a strengthening effect in healthy people¹².

Deep muscles on ultrasound

Ultrasound is a scan made with sound waves that shows soft tissue. Three ultrasound studies compared the dead bug with other exercises, and they do not agree on the transversus abdominis^2,3,4. Each measured how thick the transversus abdominis was during the exercises^2,3,4.

Three ultrasound studies compared the dead bug with other exercises, and they do not agree on the transversus abdominis.

People studied	Exercises compared	Transversus abdominis finding
44 healthy women	Five Pilates exercises, the side plank and dead bug among them	The dead bug gave a significantly higher rise in thickness, relative to rest, than the other four2
44 healthy young adults	Seven trunk exercises	The dead bug ranked fourth for relative rise in thickness, behind the bird dog, side plank and beast crawl3
13 healthy participants	Six exercises, the bird dog and dead bug among them	Thickness did not differ between the exercises4

The authors of the study of 44 women suggested a reason². In their view, some exercises need other muscle groups to work alongside the core, and the dead bug is one². They suggested these induce greater transversus abdominis activation than exercises that keep the trunk on the ground without that shared work².

The same studies reported other muscles^3,4:

• Multifidus. In the 44 healthy young adults, the dead bug ranked first of seven for relative rise in lumbar multifidus thickness, alongside the side plank³. Its rise was significantly greater than in the beast crawl and the plank³.
• Internal oblique. In the same 44 adults, the dead bug ranked behind the side plank, bird dog and beast crawl for relative rise in internal oblique thickness³. It was significantly below the side plank and the bird dog³.
• External oblique. In those 13 participants, external oblique thickness was larger in the dead bug than in the glute bridge and an exercise named the hip extension⁴.
• Transversus abdominis, internal oblique and multifidus together. In the 44 young adults, their combined relative thickness did not differ significantly between the dead bug, bird dog, side plank and beast crawl³.

The start position and the hips

One study measured the start posture of supine leg lowering, the Sahrmann test, and not of the dead bug¹³. Its 16 healthy men held the hips and knees bent to 90 degrees¹³. Holding that posture already produced baseline abdominal co-contraction, the muscles working at the same time¹³. It was about 6 to 10% of maximum voluntary contraction in the deep muscles and 8 to 16% in the superficial ones¹³.

In the dead bug form one ultrasound study used, the hips are bent to 90 degrees with the feet off the ground². Bending the thigh on the pelvis is the action of the psoas major, helped by the iliacus, two of the hip flexors¹⁴. The dead bug studies cited on this page report the abdominal muscles and the multifidus, not the hip flexors^1,3.

How the body adapts

The training trials cited on this page tested programmes of several exercises, with the dead bug as one part^6,7,8.

The training trials cited on this page tested programmes of several exercises, with the dead bug as one part.

Programmes in athletes

• 32 elite soccer players. The experimental group did six bodyweight exercises, the dead bug among them, four times a week for 8 weeks⁶. Sprint performance over 5 m and 30 m improved significantly after the 8 weeks⁶.
• 70 healthy collegiate athletes. They were randomly allocated to two groups, one of which did a five-week core stabilisation programme¹⁵. It brought in the supine dead bug in week 2, at 3 sets of 20 repetitions¹⁵. The upper quarter Y balance test and a functional throwing performance index then improved significantly, with no significant change in the control group¹⁵.

Programmes in people with low back pain

Each row below is one randomised trial in people with low back pain^7,10,16,17.

People studied	Programme that included the dead bug	What was found
48 participants with chronic low back pain	A graded lumbar stabilisation programme, one of four 6-week exercise programmes	Low back pain during physical activity fell significantly in each of the four groups7
45 participants with chronic non-specific low back pain	A 6-week core strengthening programme, against intensive dynamic back exercise	It was found more effective for functional disability and for core flexor and side bridge endurance, with no significant difference for pain10
60 people with chronic low back pain	A 4-week dynamic neuromuscular stabilization programme, against conventional core exercises	Ultrasound change rates of the transversus abdominis and the diaphragm, the main breathing muscle, were significantly greater. Pain and disability scores fell over time with no difference between groups16
58 patients with non-specific chronic low back pain	A 6-week motor control exercise programme, against manual therapy, hands-on methods such as massage or moving a joint by hand	Pain intensity and disability fell more, and the difference lasted to the 6-month follow-up17

A 2021 Cochrane review covered randomised trials of exercise in adults with chronic non-specific low back pain¹⁸. It found exercise improved pain and function compared with other conservative (non-surgical) approaches¹⁸. Those effects did not meet the review’s threshold for a clinically important difference¹⁸. Against manual therapy, it observed no difference¹⁸.

Progressive overload

Levels of difficulty

Three studies graded how hard the exercise was^5,7,8:

• 10 healthy men. The dead bug was one of four lumbar stabilisation exercises, each split into five intensity levels⁵. In general, activity of the related lumbar stabilising muscles rose significantly as the level rose⁵.
• 48 participants with chronic low back pain. A stabilisation programme had five levels from easy to difficult, each with seven positions that included the dead bug⁷. Participants started at a moderate level, and instability was increased gradually⁷.
• 43 patients with lumbar instability. A 4-week programme of five exercises, the dead bug among them, used 3 sets of 5 repetitions of 20 seconds, three times a week⁸. Participants moved to the next level when 80% of the movements were successful⁸.

Sets, repetitions, bands and surfaces

People studied	Starting dose	How the programme progressed
45 participants with chronic non-specific low back pain	2 sets of 5 to 7 repetitions per side, three alternate days a week	Over six weeks, from a knees-bent start to opposite and alternating arm and leg movements with diagonal holds10
60 people with chronic low back pain	2 sets of 10 repetitions in week 1	2 sets of 16 by week 4, with resistance bands added to the dead bug in weeks 3 to 416
32 elite soccer players	5 sets of 10 repetitions, four times a week	One set was added each week for 8 weeks6

In one randomised trial, 46 highly obese patients with lumbar instability did a 4-week progressive programme that included the dead bug¹⁹. One group used a stable surface and the other an unstable one¹⁹. Scores on a quadruple visual analogue scale, a pain rating, and on a disability index changed significantly in both groups¹⁹. The unstable-surface group alone showed a significant change in transversus abdominis thickness during contraction¹⁹.

What supine leg lowering studies varied

These findings come from supine leg lowering, the Sahrmann test, and not from the dead bug¹¹:

• Heels held up. In 22 asymptomatic men, abdominal muscle activity was significantly higher in levels 3 and 5 than in the other three¹¹. Those are the two levels where the heels did not touch the ground¹¹.
• One leg or both. In the same 22 men, abdominal muscle activity did not differ significantly between lowering one leg (level 3) and both legs (level 5)¹¹.
• Lowering speed. In 16 healthy men, one leg was lowered over 5, 3 or 1 seconds¹³. Faster lowering gave a rapid mid-movement rise in abdominal muscle activity, and slower lowering a gradual one¹³. Activity levels converged by the final phase¹³.
• Cues. In one study, 68 young adults were randomly allocated to an external cue, a pressure biofeedback unit, or an internal cue, audio²⁰. Cueing had a significant effect on activation of the abdominal muscles recorded, with no significant difference between the two cue types²⁰.

Connections

The muscles it works

These muscles, each with its own entry, appear in the studies above^1,3,4:

• The rectus abdominis, the prime mover for flexing (bending forward) the vertebral column²¹.
• The external oblique, which OpenStax names with the internal oblique as a prime mover for twisting at the waist and bending to the side²¹.
• The internal oblique, measured on ultrasound in 44 healthy young adults³.
• The transversus abdominis, on which three ultrasound studies of healthy people disagree^2,3,4.
• The multifidus, measured on ultrasound in the same 44 adults³.

Related pages

Two ultrasound studies compared the dead bug with the bird dog^3,4. Both also included a plank and a side plank^3,4.

Sources

1. Pereira ILR, Queiroz B, Loss J, Amorim C, Sacco ICN (2017). Trunk Muscle EMG During Intermediate Pilates Mat Exercises in Beginner Healthy and Chronic Low Back Pain Individuals.. Journal of Manipulative and Physiological Therapeutics. doi:10.1016/j.jmpt.2017.02.010 PMID 28413116
2. Tsartsapakis I, Gerou M, Zafeiroudi A, Kellis E (2023). Transversus Abdominis Ultrasound Thickness during Popular Trunk-Pilates Exercises in Young and Middle-Aged Women.. Journal of Functional Morphology and Kinesiology. doi:10.3390/jfmk8030110 PMID 37606405
3. Tsartsapakis I, Bagioka I, Fountoukidou F, Kellis E (2024). A Comparison between Core Stability Exercises and Muscle Thickness Using Two Different Activation Maneuvers.. Journal of Functional Morphology and Kinesiology. doi:10.3390/jfmk9020070 PMID 38651428
4. Bini RR, Bini AF (2021). Comparison of linea alba length and core-muscles engagement during core and lower back orientated exercises.. Journal of Bodywork and Movement Therapies. doi:10.1016/j.jbmt.2021.07.006 PMID 34776131
5. Kim CR, Park DK, Lee ST, Ryu JS (2016). Electromyographic Changes in Trunk Muscles During Graded Lumbar Stabilization Exercises.. PM&R. doi:10.1016/j.pmrj.2016.05.017 PMID 27317910
6. Gołaś A, Pietraszewski P, Roczniok R, Terbalyan A, Maszczyk A, Opaliński R, Zając A (2024). Effects of an 8-week pre-season targeted training on sprinting performance, agility and lower limb muscular asymmetries in elite soccer players.. Biology of Sport. doi:10.5114/biolsport.2024.134754 PMID 39416490
7. Suh JH, Kim H, Jung GP, Ko JY, Ryu JS (2019). The effect of lumbar stabilization and walking exercises on chronic low back pain. Medicine (Baltimore). doi:10.1097/MD.0000000000016173 PMID 31261549
8. Park SH, Lee MM (2019). Effects of a Progressive Stabilization Exercise Program Using Respiratory Resistance for Patients with Lumbar Instability: A Randomized Controlled Trial.. Medical Science Monitor. doi:10.12659/MSM.913036 PMID 30842392
9. Nowotny AH, Calderon MG, de Souza PA, Aguiar AF, Léonard G, Alves BMO, Amorim CF, da Silva RA (2018). Lumbar stabilisation exercises versus back endurance-resistance exercise training in athletes with chronic low back pain: protocol of a randomised controlled trial.. BMJ Open Sport & Exercise Medicine. doi:10.1136/bmjsem-2018-000452 PMID 30555717
10. Alqhtani RS, Ahmed H, Ghulam HSH, Alyami AM, Al Sharyah YHH, Ahmed R, Khan A, Khan AR (2024). Efficacy of Core-Strengthening and Intensive Dynamic Back Exercises on Pain, Core Muscle Endurance, and Functional Disability in Patients with Chronic Non-Specific Low Back Pain: A Randomized Comparative Study.. Journal of Clinical Medicine. doi:10.3390/jcm13020475 PMID 38256609
11. Chan EWM, Hamid MSA, Nadzalan AM, Hafiz E (2020). Abdominal muscle activation: An EMG study of the Sahrmann five-level core stability test.. Hong Kong Physiotherapy Journal. doi:10.1142/S1013702520500080 PMID 33005073
12. Souza GM, Baker LL, Powers CM (2001). Electromyographic activity of selected trunk muscles during dynamic spine stabilization exercises.. Archives of Physical Medicine and Rehabilitation. doi:10.1053/apmr.2001.26082 PMID 11689975
13. Yoshikawa K, Hashimoto M, Murata R, Murata Y, Jiroumaru T, Kida N (2025). Effects of different lower limb movement velocities on trunk muscle activity and dynamic stability in the Sahrmann core stability test.. PLoS One. doi:10.1371/journal.pone.0333004 PMID 40991586
14. Gray H; Lewis WH (editor) (1918). Anatomy of the Human Body. Lea & Febiger, Philadelphia (20th edition). Link
15. Jha P, Nuhmani S, Kapoor G, Al Muslem WH, Joseph R, Kachanathu SJ, Alsaadi SM (2022). Efficacy of core stability training on upper extremity performance in collegiate athletes.. Journal of Musculoskeletal & Neuronal Interactions. PMID 36458387
16. Huang H, Xie H, Zhang G, Xiao W, Ge L, Chen S, Zeng Y, Wang C, Li H (2025). Effects of dynamic neuromuscular stabilization training on the core muscle contractility and standing postural control in patients with chronic low back pain: a randomized controlled trial.. BMC Musculoskeletal Disorders. doi:10.1186/s12891-025-08417-1 PMID 40025463
17. Zhang C, Zhang Z, Li Y, Yin Y, Feng C, Zhan W, Fu R, Yu Q, Jiang G, Wang C (2024). Alterations in functional connectivity in patients with non-specific chronic low back pain after motor control exercise: a randomized trial.. European Journal of Physical and Rehabilitation Medicine. doi:10.23736/S1973-9087.24.08087-0 PMID 38358464
18. Hayden JA, Ellis J, Ogilvie R, Malmivaara A, van Tulder MW (2021). Exercise therapy for chronic low back pain.. Cochrane Database of Systematic Reviews. doi:10.1002/14651858.CD009790.pub2 PMID 34580864
19. Park SH, Lee MM (2021). Effects of progressive neuromuscular stabilization exercise on the support surface on patients with high obesity with lumbar instability. Medicine (Baltimore). doi:10.1097/MD.0000000000023285 PMID 33530156
20. Stifter S, McCaffrey J, Nichols T, Ozcan Edeer A, Ward J (2025). Effect of external and internal cues on core muscle activation during the Sahrmann five-level core stability test.. World Journal of Methodology. doi:10.5662/wjm.v15.i1.92943 PMID 40115402
21. Betts JG, Young KA, Wise JA, Johnson E, Poe B, Kruse DH, Korol O, Johnson JE, Womble M, DeSaix P (2022). Anatomy and Physiology 2e. OpenStax. Link

Every source the library cites

Changes

• October 9, 2026 First published.