“Bad posture” usually forms less from a single habit and more from repeated environments that reward the same position for long periods, especially screen work. In a randomized cross-over trial in healthy university students, laptop typing was treated as a realistic trigger for sustained upper-body postures that are commonly associated with discomfort, and the study notes that laptop use appears to contribute to poor working postures and neck pain in this population.

That framing matters: posture is not only a “willpower” issue, but often an ergonomics and repetition issue, shaped by desk height, screen position, and sustained muscle loading.

The most studied “posture problems” are forward head posture (FHP) and thoracic hyperkyphosis (an increased forward curvature in the thoracic spine). The relationship between posture and pain is not purely mechanical or one-size-fits-all, but there are measurable associations.

A systematic review and meta-analysis found that adults with neck pain showed increased FHP compared with asymptomatic adults, and FHP correlated with pain intensity and disability in adults and older adults, while the association was not found for most neck pain measures in adolescents—highlighting age as an important confounder.

On the intervention side, evidence is stronger for active, targeted approaches than for passive “hold-me-straight” solutions. A systematic review and meta-analysis on thoracic hyperkyphosis in adults reported low to moderate-quality evidence that structured exercise programs (≤3 months) reduce hyperkyphosis in younger and older populations, and low-quality evidence that bracing for ≥3 months is effective in older participants. A separate 6-month randomized controlled trial in adults ≥60 years (kyphosis ≥40°) found that a targeted spine-strengthening and posture training program produced a between-group reduction of 3.0° in Cobb angle (radiographic kyphosis) versus control, along with improvements in some clinical kyphosis measures and self-image, but no between-group differences in multiple physical function outcomes.

So where do posture correctors fit? The short-term data suggest they can change some signals, but not necessarily the outcomes people buy them for. In the scapular brace cross-over trial during a 30-minute typing task, the brace condition produced significantly smaller bilateral trapezius muscle activity, while posture changes were small and not statistically significant, and there were no immediate differences in pain or fatigue.

That pattern—measurable biomechanical shifts without immediate symptom relief—also appears in wearable biofeedback research. In adults with chronic or recurrent nonspecific neck pain aggravated by computer work, a commercially available real-time biofeedback device improved sitting posture measures and reduced cervical erector spinae activity during typing, yet self-reported neck pain was not improved in the immediate term and increased over the 1-hour task.

For low back pain, a randomized trial of soft lumbosacral orthoses over four weeks found no significant adverse effects on motor function measures overall, while disability and proprioception improved more in orthosis groups than control; increased tightness was linked to greater improvements in pain and proprioception.

The practical read is that posture devices may be useful as temporary supports or cues, especially for specific tasks, but longer-term change is more consistently supported by structured exercise and posture training, not passive correction alone.