Connecting neighboured points of categorial time series - ggplot

I have time series data: 10 second intervals (x), in each interval behavioural categories are coded (-1, 0, 1, 2, 3) (y).

I plot individuals to get a visual impression of potential patterns. So far I have the plot below and my goal is to connect points next to each other within each category of y. So that it does not look like single points but like continuous lines as long as one behavior is coded.

Here is the code for my plot so far:

ggplot(data = data_onedyad, aes(x = Sequence, y = Code_m, group = CoupleID)) +
  geom_point(colour = "blue", shape = 15) + 
  theme_minimal() +
  theme(panel.grid.minor = element_blank()

I tried with adding geom_step() but this did not lead to the desired result (because it connects all points across categories of y):

You can identify sequences of the same values with run length encoding. vctrs::vec_identify_runs(), dplyr::consecutive_id(), or data.table::rleid() return to what run a value belongs, which makes a good grouping variable in your case.

library(ggplot2)

df <- data.frame(
  x = factor(1:20),
  y = sample(c(-1, 0, 1), 20, replace = TRUE)
)

ggplot(df, aes(x, y)) +
  geom_point() +
  geom_line(aes(group = vctrs::vec_identify_runs(y)),
            colour = "red")

<!-- -->

^{Created on 2023-04-19 with reprex v2.0.2}