I want to draw a bar chart with three columns, where two arrows start from one column to the other two columns and these arrows do not overlap each other, as shown below.

I managed to produce the chart and I used geom_segments to draw the arrows. However, the arrows overlap when they start from the column control.

I thought I could handle the problem by setting geom_segment(..., position = position_nudge(x = 0.25)). However, this changes not only the position of x but also xend. Then, how should I slightly change geom_segment's position of x only, keeping the position of xend constant?

MWE

library(tidyverse)

demoData <- tribble(
  ~priming, ~rt,
  "control", 374,
  "phonological", 267,
  "orthographic", 304
) |>
mutate(
  diff.from.baseline = rt - first(rt)
)
  
baseline <- demoData |>
  filter(priming == "control") |>
  dplyr::select(rt) |>
  pull()

demoData |>
  ggplot(
    aes(
      x = priming,
      y = rt
    )
  ) +
  geom_col() +
  #### from control to orthographic
  geom_segment(
    aes(
      x = "control",
      y = baseline,
      xend = "control",
      yend = baseline + 50
    )#,
    #position = position_dodge2(
    #  width = 0.5,
    #  preserve = "total"
    #  )
    #position = position_nudge(
    #  x = 0.25,
    #  xend = 0
    #)
  ) +
  geom_segment(
    aes(
      x = "control"  + 0.25,
      y = baseline + 50,
      xend = "orthographic",
      yend = baseline + 50
    )#,
    #position = position_dodge2(
    #  width = 1,
    #  preserve = "total"
    #  )
    #position = position_nudge(
    #  x = 0.25,
    #  xend = 0
    #)
  ) +
  geom_segment(
    aes(
      x = "orthographic",
      y = baseline + 50,
      xend = "orthographic",
      yend = demoData |>
        filter(priming == "orthographic") |>
        dplyr::select(rt) |>
        pull()
    ),
    arrow = arrow()
  ) +
  #### from control to phonological
  geom_segment(
    aes(
      x = "control",
      y = baseline,
      xend = "control",
      yend = baseline + 100
    )
  ) +
  geom_segment(
    aes(
      x = "control",
      y = baseline + 100,
      xend = "phonological",
      yend = baseline + 100
    )
  ) +
  geom_segment(
    aes(
      x = "phonological",
      y = baseline + 100,
      xend = "phonological",
      yend = demoData |>
        filter(priming == "phonological") |>
        dplyr::select(rt) |>
        pull()
    ),
    arrow = arrow()
  )

Here's an approach:

segments <- data.frame(seg = rep(c(1:2), each = 4),
                       x = c(0.8, 0.8, 3, 3, 1.2, 1.2, 2, 2),
                       y = c(as.numeric(demoData[1,2]), 450, 
                                        450, as.numeric(demoData[3,2]),
                            as.numeric(demoData[1,2]), 425, 
                                       425, as.numeric(demoData[2,2])))

ggplot() +
  geom_path(data = segments, aes(x, y, group = seg), arrow = arrow()) +
  geom_col(data = demoData, 
           aes(x = as.numeric(factor(demoData$priming, 
                                     levels = demoData$priming)), rt)) +
  scale_x_continuous(breaks = 1:3, labels = demoData$priming)