# Lower Trapezius Exercises: Don't Train the Sore Spot > Updated: 2026-06-10 · Source: https://dorsi.ai/blog/lower-trapezius-exercises-neck-pain One trial cut chronic neck and shoulder pain with two exercises that avoid the muscle that hurts. What it trained, and what that does not prove. ![An illustration of a person on a bench doing a Y-raise with light dumbbells, seen from behind, with a soft green bloom across the middle and lower back rather than the top of the shoulders.](/blog/lower-trapezius-exercises-neck-pain.webp) When the top of your shoulder aches after a long week at a keyboard, the instinct is to go straight at it. Dig a thumb into it, roll it on a ball, book a massage, and if you train, throw in some shrugs to toughen it up. There is a trial worth knowing about that went the other way on purpose. It built a programme around the muscles below and around the shoulder blade, and deliberately kept direct work off the part that hurt.

Key Takeaways

## The trial that trained around the pain In 2014, a group at Denmark's National Research Centre for the Working Environment published a [randomized trial in the *Journal of Occupational Rehabilitation*](https://pmc.ncbi.nlm.nih.gov/articles/PMC4000422/) with an unusual design. They recruited forty-seven office workers with chronic non-specific pain in the neck and shoulder region, and gave half of them ten weeks of scapular function training: three sessions a week, twenty minutes each. The unusual part is what the programme targeted. It trained the lower trapezius and the serratus anterior, the muscles that sit below and around the shoulder blade, while deliberately minimising direct training of the upper trapezius. The upper trapezius is the muscle most of these people would have pointed to if you asked them where it hurt. Over ten weeks, pain intensity in the training group fell by 2.0 points on a scale the researchers judged clinically relevant. Pressure pain threshold in the lower trapezius, which is a measure of how much force it takes before pressure becomes painful, rose by 129 kPa. Shoulder elevation strength went up by 7.7 kg. Forty-seven people in one country is a small trial, and the participants were office workers rather than a general population, so this is a direction rather than a guarantee. ## The model the design came from The reasoning behind that design is older than the trial. Physiotherapists have described a pattern since Vladimir Janda's work, [usually called upper crossed syndrome](https://pubmed.ncbi.nlm.nih.gov/41595914/), in which the upper trapezius sits on the tight and overactive side while the middle and lower trapezius sit on the weak side. I have laid the full map out in the [piece on forward head posture](/blog/tech-neck-forward-head-posture-fix); the part that matters here is just that one line. In that model the sore muscle is on the overactive side and the underworked ones sit below it, which is the reasoning that led someone to design a programme avoiding the painful area. Whether that model is true of your shoulder is a separate question the trials did not ask. That is the logic the 2014 programme was built on. It is worth being clear about what it does and does not establish. The model comes from a narrative review describing a typical clinical picture, not evidence that the upper trapezius causes any given person's pain, and the trial never tested direct upper trapezius training against the scapular programme. What was shown is narrower: a programme built to minimise direct upper trapezius work reduced pain compared with no training. ## Active beat passive, head to head Most of the evidence in this area compares an exercise programme against doing nothing, which leaves an obvious question. Is the exercise doing something specific, or would any attention to the area help? A [2022 randomized trial in *Medicine*](https://pubmed.ncbi.nlm.nih.gov/36181044/) put that to the test directly. Forty-one people with chronic neck pain were randomly assigned either to a cervical and scapula-focused resistance exercise programme or to trapezius massage, five times a week for four weeks. Both groups improved within themselves, which is worth saying plainly, because massage is not useless. On the between-group comparison the resistance exercise group came out significantly ahead on pain, cervical rotation, upper trapezius tone and stiffness, disability scores and quality of life. That is a small trial over four weeks, and it cannot tell you what either approach does over a year. It also cannot isolate why exercise did better, since the two interventions differ in more than load: movement, effort, expectation and the structure of the programme all vary together. ## What this looks like in a session It is worth knowing how narrow the tested programme actually was, because most articles on this subject hand you a list of ten movements. The 2014 trial used two exercises. A push-up plus, which is a push-up that finishes by pushing the shoulder blades apart at the top, and a press-up, where you sit and press down through your hands to lift your weight off the seat. Both were chosen because earlier electromyography work showed they activate the serratus anterior and lower trapezius strongly while leaving the upper trapezius comparatively quiet. Resistance was added with elastic bands across the back or over the shoulders, and the load progressed from roughly a 20-repetition maximum in the first week to a 10-repetition maximum by the last, three sets of each. That is the whole intervention that produced the result described above. Two movements, ten weeks, progressively loaded. One detail about setup, with its limits attached. An [EMG study published in 2023](https://pubmed.ncbi.nlm.nih.gov/37425105/) found that shoulder blade position changed which muscles did the work during a side-lying abduction exercise. That study measured twenty collegiate baseball players holding isometric positions, not desk workers with neck pain, and it measured muscle activation rather than whether anyone felt better. It is a reason to pay attention to where your scapula sits, not a coaching cue the evidence has validated for pain. **When to see someone instead.** Everything here concerns ordinary non-specific neck and shoulder pain from desk work. Pain that radiates down an arm, follows an injury, or comes with numbness or weakness is a different problem and needs a clinician. ## What this is worth Two small trials in office workers is a thin base to build a routine on, and neither tested the obvious alternative of training the sore muscle harder. What the 2014 result does show is that two exercises, neither of them aimed at the painful area, were enough to move chronic pain in ten weeks. Whether that beats the thing you would have done instead is a trial nobody has run. ## References 1. Andersen CH, Andersen LL, Zebis MK, Sjøgaard G. [Effect of scapular function training on chronic pain in the neck/shoulder region: a randomized controlled trial](https://pmc.ncbi.nlm.nih.gov/articles/PMC4000422/). *Journal of Occupational Rehabilitation*. 2014;24(2):316–324. doi:10.1007/s10926-013-9441-1. Free full text. 2. Kang T, Kim B. [Cervical and scapula-focused resistance exercise program versus trapezius massage in patients with chronic neck pain: A randomized controlled trial](https://pmc.ncbi.nlm.nih.gov/articles/PMC9524908/). *Medicine* (Baltimore). 2022;101(39):e30887. doi:10.1097/MD.0000000000030887. Free full text. 3. Russin NH, Robertson C, Montalvo A. [Upper Crossed Syndrome in the Workplace: A Narrative Review with Clinical Recommendations for Non-Pharmacologic Management](https://pubmed.ncbi.nlm.nih.gov/41595914/). *International Journal of Environmental Research and Public Health*. 2026;23(1):120. doi:10.3390/ijerph23010120. 4. Tsuruike M, Ellenbecker TS. [Effect of Scapular Retraction on Lower Trapezius, Infraspinatus, and Deltoid Muscle Electromyographic Activity During the Side-Lying Abduction Exercise](https://pubmed.ncbi.nlm.nih.gov/37425105/). *International Journal of Sports Physical Therapy*. 2023;18(3):715–725. doi:10.26603/001c.74969.