Abstract Most swine production systems in the US do not include access to environmental enrichment: an improvement in the biological functioning of captive animals resulting from modifications to their environment. Optimal enrichments such as straw, hay and silage are not acceptable due to manure system and biosecurity concerns. There remains a need to develop enrichment for pigs that is biologically relevant, safe for existing systems, biosecure, sustainable and cost effective. The aim of this study was to compare the behavior and welfare of sows over 2 weeks post-mixing when given access to a suspended molasses block (B - Barastoc Swine Block) or a metal chain (C). Three replicates of 36 LW × Yorkshire sows, balanced by parity, were mixed at d 42 post service in groups of 4, 6 or 8, with one of each group size per treatment. Groups were formed with equal numbers of sows from 2 large group pens, i.e., a group of 4 was formed using 2 sows from one pen, mixed with 2 sows from a second pen, and so on. Sow behavior and animal-based welfare measures were recorded 3 days before mixing, and d 1, 6 and 13 post-mixing. Welfare measures included tear stains, lesion scores, bursitis, shoulder sores, respiratory and cleanliness scores. Behavior included passive and active human-approach tests, where the human stood for 3 min against one wall of the pen to allow sows to approach, noting number to approach and touch the human and the latency to touch. Then at the end of the test, the human actively approached each sow in turn and touched the head of the sow, noting the response of the sow on a numerical scale. Data were analyzed using 2-way ANOVA with treatment and group size as factors. There were no treatment differences in bursitis, shoulder sores and respiratory scores, and no differences in time to approach and touch the human. However, B sows had smaller tear stains (d 6 and d 13, P < 0.001), decreased body lesion scores (d 1 and d 6, P < 0.05), and better body cleanliness scores (d 1, P < 0.05). Human approach test measures showed habituation over time, but B sows responded less to a human approaching and touching their head than C sows (d 6, P < 0.05). Groups of 6 had lowest tear stain scores (d 1, d 6 and d 13, P < 0.05) and were cleanest (d 13, P < 0.05). There was also a Group size*treatment interaction with groups of 6 having the lowest lesion scores in the C treatment, but the highest in the B treatment (d 1, d 6, and d 13, P < 0.05). The skin lesions scores indicate that aggression was lower in the pens with the blocks. Overall, the blocks improved sow welfare over the immediate 2-week post-mixing period, though group size may impact this.